JOINT RESEARCH



PAKISTAN-US SCIENCE AND TECHNOLOGY

COOPERATIVE PROGRAM 2007

JOINT PROJECT PROPOSAL

• Please provide the required information briefly. Condensed statements are preferred. However, the boxes in Section 3 may be expanded as necessary. The maximum limit of the project document is 10 pages.

• Please do not forget to attach the one-page curricula vitae of all the participating scientists.

|Project title: |

|Measurement and Analysis for the Global-grid and PERN’s |

|Internet end-to-end Performance (MAGPIE) |

|Field and subfield of proposed project |

|Internet Communications: Network Performance Measurement and Analysis |

|Duration: 3 Years |Starting date: January 2008 |

|PAKISTAN SIDE |US SIDE |

|Principal Investigator: Prof. Dr. Arshad Ali |Principal Investigator: Dr. Les Cottrell |

|NUST Institution of Information Technology |Stanford University/SLAC |

|Date |Signature |Date |Signature |

| | | | |

|Address |Address |

|166A, Street 9 Chaklala Scheme III |Stanford University/Stanford Linear Accelerator Center |

|Rawalpindi Pakistan 46000 |2575 Sand Hill Road, MS 96 |

| |Menlo Park, CA 94025 |

|Telephone |Fax |E-mail |Telephone |Fax |E-mail |

|92-51-9280433 |92-51-9280782 |arshad.ali@niit.edu.pk |(650) |(650) |cottrell@slac.stanford.edu |

| | | |926-2523 |926-3329 | |

|Head of Institution/Vice-Chancellor/Rector |Head of Institution/Vice-Chancellor/Rector |

|Lt. General (R) Syed Shujaat Hussain, Rector, |c/o Meredith O’Connor, |

|National University of Sciences and Technology |Managing Senior Contract and Grants Officer |

|Tamiz-Ud-Din Road, PO BOX 297 |Stanford University |

|Rawalpindi, Pakistan |Office of Sponsored Research |

|Email: rector@nust.edu.pk |320 Panama Street |

|Phone: +92-51-9271575, +92-51-9271576 |Stanford, CA 94305 |

|FAX : +92-51-9271577 |e-mail: meredith.oconnor@stanford.edu |

2. Project Summary

|Title of the Project: Measurement and Analysis for the Global-grid and PERN’s |

|Internet end-to-end Performance (MAGPIE) |

|Pakistan Side |

|Co-Director: Prof. Dr. Arshad Ali |

|Institution: NUST Institution of Information Technology (NIIT) |

|U.S Side |

|Co-Director: Dr. Johnathan Dorfan |

|Institution: Stanford University/Stanford Linear Accelerator Center (SLAC) |

| |

|The growth of the Internet has fostered intellectual and economic growth of both developed and developing countries [[i]]. Fundamental to this is the |

|operation and maintenance of the underlying Internet technologies to keep the network functioning and performant. |

|Specifically, advancements in network connectivity have improved scientists’ ability to collaborate and helped accelerate the rate of scientific |

|discovery, see for example [[ii]]. Modern science relies on the global Internet to create large and physically distributed scientific work environments.|

|Thus scientists have become more dependent on networking, and network problems have increasing significance. The pursuit of modern science is thus |

|dependent upon cyber-infrastructure designed to support the efficient diagnosis of network issues along end-to-end paths traversing multiple networks. |

|We outline a proposal to enable Pakistan’s advanced education and research facilities to better understand and utilize their network connections and for|

|Pakistan’s Education and Research Network (PERN) to become a world leading National Research and Education Network (NREN). Viewing case studies such as |

|[[iii]] it is apparent that by monitoring the network, even for Pakistan alone, one can identify the fragility of computer networks; the problematic |

|routing of traffic e.g. between sites in Islamabad and Rawalpindi going via London; and despite well provisioned backbones, last mile problems causing |

|heavy congestion and hence back end-user experience of the Internet. |

|Understanding network performance can also pose challenges for distributed collaborators. Scientific communities usually span many physical |

|organizations. The distributed nature of these communities makes it difficult for campus network engineers or end-users to effectively find or fix |

|end-to-end performance problems. Campus networks are usually interconnected via regional networks and national backbones and may span many domains. Each|

|of these domains usually maintains explicit control over who can access information and infrastructure and what performance data is shared, if any. |

|Often, considerable time and effort must be expended to convince physical organizations to begin testing and it can still be a difficult task to isolate|

|the issue, and even more time can be spent evaluating different solutions. It is therefore difficult to perform tuning and troubleshooting in |

|distributed environments due to the need to access information from many different sources and the human coordination required. |

|Hence, it is critical that any solution to the problem of network performance monitoring and diagnostics must be scalable and universal to enable |

|investigators to easily gather network performance data. More importantly, the framework for this should be capable of conducting automatic analysis, |

|diagnosis and reporting of network problems for different types of users (e.g. end-users, backbone administrators etc.). |

|SLAC, Internet2, ESnet, GEANT, University of Delaware, Georgia Tech and RNP (Brazil) have begun a global project to deploy such monitoring |

|infrastructures under the perfSONAR framework [[iv]]. We propose a collaborative effort between SLAC/US and NIIT/Pakistan to help further develop the |

|various tools and components that will facilitate perfSONAR and apply these tools to the new PERN network including major end-sites. We intend to help |

|build core services and infrastructure to facilitate the continued global support of perfSONAR and enable successful scientific collaborations between |

|Pakistani scientists and their peers worldwide. |

|We will also tutor and supervise students at NIIT and offer select graduates the opportunity to work in the US, where they will gain knowledge of |

|research and development in the domain of network performance monitoring as well as valuable experiences of working at a world class leading research |

|institute, contacts and best practices in many fields, that they will take back to Pakistan. |

3. Project Description

|3(a) Background and Rationale |

|To efficiently manage any network one needs to be able to measure it. This includes measuring and understanding current and long-term performance, |

|identifying and reporting problems both end-to-end and within the network itself, and providing forecasts of both long and near term performance. |

|Without such information network managers lack planning information, users and network administrators do not know what to expect; problems are reported|

|by users, and network administrators spend all their time fire-fighting; locations and sizes of bottlenecks and their behavior are unknown; and users |

|and applications are unable to dynamically optimize their network usage. |

|The extent of ensuring that the network ‘works’ is demonstrated by the fact that network researchers and end-users have always had a need for network |

|measurement, monitoring, and analysis tools. This need dates back to the early days of the Internet when the “ping” command was written. While tools |

|like the ubiquitous “ping” and “traceroute” commands will continue to be the front-line tools used by most network engineers, more sophisticated tools |

|are needed to directly measure one way delays, throughput and bandwidth, utilization, non layer 3 paths, etc. Network monitoring tools and how they are|

|used should answer both basic connectivity type questions and more complex “what’s wrong with this path” type questions. |

|Previous efforts to develop these new tools can be sorted into two major classes: (1) individual tools that can be installed on Internet hosts, (2) |

|measurement tools deployed in a single test domain. |

|The first class of projects deals with the development of individual measurement tools. These refer to two types of measurement tools: passive and |

|active. Passive tools (e.g., tcpdump [[v]], OCx-mon [[vi]], Netflow [[vii]], and Web100 [[viii]]) make use of application-generated data packets that |

|flow through the network. A copy of the data packet is passed to the tool for some type of processing. Each tool typically generates its own unique |

|output format. As with active tools (described below), the output of the tool is usually handed over to a tool expert for interpretation and analysis. |

|Active tools (e.g., Iperf [[ix]], Thrulay [[x]], NDT [[xi]], NPAD [[xii]]) use a client/server model to generate test packets. The major advantage of |

|these tools is that individuals and small communities can download and install the tools on an as needed basis. This ensures that tools are used in a |

|wide variety of operating environments, much larger than the developers could have achieved. A major disadvantage is that these tools may not be |

|already installed when problems are reported. Another deterrent is that remote administrators may be reluctant to install them when they are not sure |

|how the tool works or what its reporting capabilities are. Finally, tools that require manual action at both ends usually require coordination between |

|remote administrators and an implicit trust relationship, reducing ease of use. |

|The second class of projects deploy network-only measurement or monitoring tools inside some test domain. In addition to individual tools, larger |

|research projects (such as AMP [[xiii]], Surveyor [[xiv]], IEPM-BW [[xv]], PingER [[xvi]], RIPE’s Test Project [[xvii]]) have been established to |

|generate large-scale measurement and monitoring programs. These projects tend to use both active and passive test tools to perform their measurement |

|and monitoring functions. A major disadvantage to such systems developed in such projects is that they have lacked community involvement in the design |

|and are ultimately limited by the ability of the research team to support such an infrastructure. As a direct result, such complete systems are hard |

|for non-project members to extend as they have lacked community involvement in their design. Experience has shown that such projects (see for example |

|AMP and Surveyor) fade away as funding shifts and people move on to new projects, and those subsequent projects rarely build on what has come before. |

|Therefore, a single, monolithic, global measurement system is unrealistic. Federation and neutral information exchange standards-based formats and a |

|modular mix-and-match architecture are necessary. This preserves local autonomy while promoting global utility. Moreover, reports and unified access to|

|the measurement results are needed to provide useful analysis and diagnostics in order to identify, diagnose and help resolve network performance and |

|connectivity problems to ensure that users can view the network as a reliable, predictable utility. |

|3(b) Problem Statement |

|The next generation of the PERN backbone, PERN2 [[xviii]], will soon be deployed to interconnect over 60 universities across Pakistan and to the rest |

|of the world. It will offer immense possibilities for the various network users across diverse disciplines/endeavors such as medical research [[xix]] |

|[[xx]], particle physics [[xxi]], agro-informatics [[xxii]], media-communications [[xxiii]] [[xxiv]] and distance learning [[xxv]]. |

|Figure 1 shows the historical connectivity of the existing international link between SLAC and NIIT. As the figure shows, network users typically |

|experience very poor performance (RTTs > 250ms and very variable and heavy losses) and network fragility (100% loss) that makes collaborative |

|technologies such as video conferencing technically unfeasible at least until late 2006 when performance improved. As PERN2 is deployed, it is |

|important to not only understand but also to monitor the network infrastructure and the end-to-end performance to ensure that the return on investment |

|of such a large project is understood and that network users are able to utilize such infrastructures. |

|[pic] |

|Figure 1: PingER RTT (blue) in msec. and loss (green) % since May ‘05 thru May ‘07 measured from SLAC to NIIT |

|To understand the problems of network performance and monitoring, and their effect on network users, the research community and the local industry, it |

|is beneficial to note how these groups typically deal with network performance measurement and monitoring issues these days. |

|Consider two scholars from different universities (but connected by the PERN network) arranging a video conference to facilitate their collaboration. |

|Hypothetically, when they setup the call, one scholar finds that the video stream is poor. The scholar then contacts his/her local network |

|administrator seeking help in resolving the matter. After some preliminary tests the network administrator suspects that the reason for the poor video |

|quality is insufficient available bandwidth. For this reason he/she plans to use the Iperf tool to test the links. However, in order to use Iperf there|

|is a requirement of a server component. Consequently, after hours or possibly days of telephone tag and coordination with network administrators within|

|each administrative domain, the tests are conducted and a single link identified behaving as the bottleneck. Later the network administrators |

|corroborate to fix the problem and scholars are informed that the link would be available after the repair has been made. This is all with the |

|assumption that the network administrators within each administrative domain forming the link were available and willing to support the tests. Thus in |

|such scenarios it is safe to say that days of effort from multiple network administrators is required to first identify the problem, and later fix it. |

|Reducing the hours and number of participants required to find and report these types of problems is a major goal of this proposal. Not only is the |

|time to repair reduced, but the frustration level of the scientist is also reduced. Knowing where a problem is, and when it is likely to be repaired, |

|can help scientists develop contingency plans. In addition to making manual trouble reporting and resolution easier, a network measurement |

|infrastructure can assist network operators with the day-to-day monitoring tasks that can automatically find and report problems. |

|Waiting for trouble reports from users before noticing a problem is another serious issue with today’s complex network. While a typical operations |

|staff quickly notices and fixes a major outage or a problem with a well-used application, a niche connection or application may fail for weeks or |

|months before a trouble report makes it into the Network Operational Center (NOC). See for example [[xxvi]] where dramatically reduced throughputs went|

|un-noticed and un-reported for more than two months, yet when reported took only hours to fix. In the mean time, the end-user is frustrated and unable |

|to effectively use the network to perform a specific task. |

|At a higher level, the impact of this work on society will mirror that of the Internet itself; by making networks easier to operate and use, scientists|

|and end-users alike will have an improved experience. This translates to scientific and commercial benefits from the ability to more effectively use |

|existing applications or take advantage of new ones [[xxvii]]. |

|Today, numerous scientific disciplines have developed globally distributed work environments that depend on the network. Some examples include: |

|The international High Energy and Nuclear Physics (HENP) community is nearing completion of the Large Hadron Collider (LHC) [[xxviii]] facility located|

|at CERN in Switzerland. The LHC will allow thousands of scientists, engineers, and university professors access to petabytes of particle physics data. |

|Data will be stored at multiple national repositories and processed at dozens to hundreds of sites around the globe. Both NUST and the National Centre |

|for Physics (NCP) at Quaid e Azam University and others are building computer clusters to be used for simulations and analysis for LHC and others. |

|Without highly reliable and high performance network connections to CERN, Europe and the US, these clusters will have limited global impact as they |

|will be unable to replicate and transfer the large amounts of data required for simulation and analysis. |

|Cisco (Mr. Horst Dumcke and Dr. Masum Hassan) in collaboration with NIIT (Dr. Arshad Ali) has been aggressively pursuing the establishment of a ‘Health|

|e-Grid’ [[xxix]]. Due to the fact that sufficient trained medical professionals are not available in Pakistan (even in urban areas), providing health |

|facilities to the population (of 160 million with a growth rate of 2.09%) in general is a challenging task. The intention behind the Health e-Grid |

|project is to setup a tele-medicine infra-structure so as to provide basic medical facilities to the population in remote areas. A step towards |

|providing affordable health care to Pakistani public is to make use of Grid technologies to provide the facility of telemedicine. Given the critical |

|nature of such Grid applications, the setup, management and maintenance of PERN – the network over which the applications are intended to execute – is |

|of vital importance. |

|The Virtual University [[xxx]] of Pakistan delivers education through a judicious combination of broadcast television and the Internet. The lectures |

|and miscellaneous literature is made available over the PERN in the form of streaming media and documents/archives from the Virtual University’s |

|servers. Here too the predictable behavior of the underlying network is desirable which can only be achieved if it is suitably monitored, the network |

|is reliable and anomalies are either identified in time to limit major disruptions.. |

|Organizations such as Pakistan Agricultural Research Council, Agriculture University Faisalabad, Agriculture University, Peshawar have been |

|collaborating under the Agro-Informatics [[xxxi]] initiative and have achieved significant results. Such initiatives have only been possible due to the|

|availability of means to conduct collaborative research. Existence of networks such as PERN encourage such collaborations by providing suitable means |

|for communications among the collaboration. |

|On a more technical level, the rapid detection and correction of network performance problems and infrastructure faults represent unresolved problems |

|facing all these scientific communities. It is difficult to overstate the amount of effort that large-scale distributed projects such as these spend on|

|network debugging and performance tuning. That this effort is duplicated across projects represents a significant waste of time and energy. Moreover, |

|many scientists are forced to become part-time network engineers, which is an inefficient use of their time and effort. |

|Therefore, a network monitoring infrastructure should focus on the following: |

|Implement techniques to share network performance data across domains to enable a broader snapshot of network performance with the end-to-end view. |

|Enable and accommodate new types of analysis that are continually being developed. Sometimes, these are done using completely new tools that generate |

|information on completely new metrics. Often, they represent new ways of correlating existing diagnostic data types. |

|Be easily adaptable to changes in network technology to support new devices and provide useful diagnostic data. This may include ease of extensibility |

|to new data sources – new tools are produced frequently and the system must not be biased toward any particular tool. |

|Secure, policy-based access: the ability to provide protection for who can access what and under what conditions at a local level is critical for a |

|single system framework to be accepted. |

|Community-driven: the architecture of the system must be completely open (and open source) to encourage the reintegration of extensions and the |

|piecemeal upgrading of modular components. |

|Another aspect of network performance is that Network Operational Centers (NOCs) typically are focused on monitoring the performance of their own |

|network domains. However, real Internet users will have to traverse many different domains in order to do their work. As such, the typical end-to-end |

|performance experienced by real network users, is often nowhere near that which NOCs would assume. Example limits in the performance of end-to-end |

|experiences include: |

|Badly tuned network components (including mismatched speed or duplex issues, stack limitations or even TCP algorithmic limitations). |

|Routing problems; including mis-configured routers and switches; non-optimal routes; poor backup routes; flapping etc; all causing unpredictable |

|performance. |

|Overloaded networks which have been badly provisioned to support the traffic that upon it. |

|Denial of Service attacks; maintenance; power problems; firewall limits; port blocking; congestion; improper host configurations; application problems.|

|Real-life examples of all of these can be found in the SLAC Case Studies referenced above. |

|3(c) Prior Experience / Capability |

|The perfSONAR consortium initiated in partnership by Internet2 [[xxxii]], the European Union-funded GÉANT2 project [[xxxiii]], and DoE Energy Sciences |

|Network (ESnet) [[xxxiv]], has been working, in recent years, to provide an open set of protocols and a reference implementation to address |

|cross-domain sharing of network measurements and metrics. SLAC recently joined the consortium - providing their experience and skills in deploying the |

|PingER and IEPM-BW network monitoring infrastructures; their expertise in network data mining and analysis; as well as their close involvement with |

|HENP [[xxxv]] and the LHC [[xxxvi]]. |

|The perfSONAR project is focused on building open source, flexible, modular, and extensible performance middleware to greatly simplify the process of |

|gathering and sharing network performance information across multiple administrative domains to simplify the process of debugging network issues and to|

|provide performance data to network-sensitive applications without redundant measurements. Early prototypes of the perfSONAR framework are deployed in |

|Europe and in North and South America and serve as a central component of the network monitoring functionality for the global network dedicated to |

|serving the LHC project. The rapid adoption of the early prototype of the perfSONAR system by various NRENs and the LHC project is a testament to the |

|need that it fills. |

|The perfSONAR project originated with the desire to unify the efforts of Internet2’s End-to-End Performance Initiative Performance Environment System |

|(E2E piPES) project [[xxxvii]] and the EU’s GÉANT2 Joint Research Activity on monitoring and measurement (GN2-JRA1) [[xxxviii]], ultimately involving |

|parallel efforts in the U.S. DoE’s ESnet. The desired result of this project, a common performance middleware, is referred to as the perfSONAR |

|framework. This project seeks to enable consistent access to network performance data across a large fraction of the world’s advanced research and |

|education networks. The effort is in its second year of funding in the EU and receives a small amount of seed money from Internet2 and ESnet in the |

|United States. Some components of this framework exist in prototype form, but much remains to be done. |

|This framework specifically defines a federated network measurement infrastructure that can be used by a large number of research scientists and |

|scientific collaborations. The perfSONAR project is still at a very early phase, with much work to be done yet, but it is beginning to have a |

|significant impact already and to generate interoperable, independent development of additional services, which is a testament to the level of |

|community support. |

|perfSONAR defines a series of mid-level services that can be used to create a measurement infrastructure. It does not mandate the use of a specific |

|tool or set of tools to be used to generate or capture measurement data. Rather, it defines a standard XML-based storage schema that allows multiple |

|tools to generate and consume data. This increases the usability and reach of these tools, making them more applicable to a larger number of potential |

|users. perfSONAR does not require real-time intervention to deploy tools or to coordinate the data collection process. Instead, a standard |

|authentication mechanism allows interested parties (e.g., end-users, scientists, dedicated support staff, network operators) to pre-register themselves|

|in a domain-managed database. Once registered, these users can present their credentials to a remote domain’s federated authorization service and, if |

|approved (according to locally-determined policies, and home institution defined identity), they can begin troubleshooting a problem. |

|In summary, the perfSONAR framework will ultimately: |

|Be easily extensible to new types of measurement and analysis tools. Tool developers will be able to focus on how the tool operates, not on how to |

|deploy the tool widely enough for it to become useful. |

|Scale to global size. Multi-national science communities will be able to focus on science, not just run networks. |

|Contain flexible policy for sharing across administrative domains. Local domains will retain control over the policies that define how specific |

|resources can be accessed and what data will be shared with whom according to uniform, organization-agreed-upon policies, not ad hoc decisions based on|

|personal relationships. |

|Use common data exchange formats. The OGF [[xxxix]] NM-WG [[xl]] schemata will be used to ensure that different tools and archives can exchange data |

|effectively. |

|Use common discovery mechanisms. Users will be able to find measurements points and archival data to assist in troubleshooting an application’s |

|behavior. |

|Use generic service components for tracking resource utilization and user authorization to acquire those resources for performance testing. |

|Piggyback on existing federated trust authentication infrastructures. |

|TWe believe the perfSONAR framework provides the necessary services and infrastructure to facilitate a rich and extensible monitoring system for PERN2 |

|from which further network research and opportunities can arise. |

|Various funded projects in the past have failed to reach critical mass in the realm of network monitoring due to their lack of open source and open |

|community traits. Furthermore, projects, such as AMP, MonALISA [[xli]], IEPM-BW, PingER and Surveyor, have attempted to bundle data generation, data |

|sharing, and analysis with visualization in an attempt to do everything. However, they have been hampered by the relative lack of standards for sharing|

|performance data, limiting the breadth of data (both in type and deployment) available to be consumed. Many members of the perfSONAR development team |

|have been directly involved with creating or using these tools and the lessons learned during those developments have been incorporated into the |

|perfSONAR framework. |

|SLAC has been a leader in network monitoring and performance tuning for many years and has over fourteen years of computer networking experience in the|

|wide area with experiences ranging from the deployment of the first Internet in China [[xlii]], capturing the Internet2 Land Speed Record twice |

|[[xliii]], winning the Supercomputing Bandwidth Challenge three years in succession (2003, 2004 and 2005 [[xliv]]), to the monitoring of the digital |

|divide between developed and developing countries [[xlv]]. SLAC also has extensive experience developing and deploying network infrastructures such as |

|the PingER suite and the IEPM-BW suite – both of which are still collecting data on over 125 countries from over 40 sites throughout the world. |

|With the benefits of decentralized network monitoring provided by the perfSONAR framework, SLAC has shifted its focus from the deployment and |

|collection of metrics to the analysis and presentation of network performance. An important aspect to the analysis of network performance data is the |

|understanding of the baseline performances and the effects of seasonal effects such as the increase of network traffic during work days and hours. |

|This is illustrated in the figure below for throughput measurements from SLAC to the University of Florida, where it is seen (blue line) that at the |

|weekend measured performance is higher and then oscillates (day down, up at nights) during the week. The attempt to use an exponentially weighted |

|moving average (light blue smooth line) to predict the performance lags the data and, following the weekend, the differences are outside the acceptable|

|range (below the orange line) for a sufficiently long duration (the black line is the number of events falling outside the threshold) that an event is |

|detected. |

|[pic] |

|Figure 2: Detection of anomalous events using the plateau algorithm [REF] from SLAC to U. Florida |

|The net result is that there is a higher probability for detecting false events at the end of a weekend or the start of a weekday. This is illustrated |

|in the plot below where we see the number of events triggered by the simple non-seasonal Kolmogorov-Smirnov [[xlvi]] algorithm is weighted towards |

|21:00 hours Pacific time (which is 9:00am Pakistan time, or the start of the work day when utilization increases and congestion starts to get bad).  |

|[pic] |

|Figure 3: Frequency of detecting events as a function of time of day using the Kolmogorov-Smirnov algorithm on available bandwidth between SLAC and |

|NIIT. |

|Thus for paths with seasonal variations we need to investigate more sophisticated techniques including Holt-Winters [[xlvii]] and ARMA/ARIMA [[xlviii]]|

|models. The authors of this proposal have built prototypes using Holt-Winters [[xlix]] and investigated ARMA/ARIMA [[l]] for forecasting and event |

|detection. An example of the use is shown in Figure 4. Given the noisiness of the data (dots) it is first smoothed (blue line), Holt-Winters is then |

|used to forecast the performance (red line of bandwidth capacity) and provide expected deviations (black lines). If the smoothed data goes outside the |

|deviations for sufficient number of points then an event is signaled. It is seen that the Holt-Winters forecast tracks the seasonality of the data, and|

|detects the sudden and persistent drop in performance on Friday evening/Saturday morning July 31, 2004. |

|[pic] |

|Figure 4: Time series of bandwidth capacity measured between SLAC and U. Indiana with forecasting by Holt-Winters, and the detection of an event. |

|The above is a simple example of the types of analysis that network engineers and end-users would analyze to diagnose a network problem. As the |

|perfSONAR framework enables the collection and storage of results in a simple and consistent format, it offers new opportunities that will lead to a |

|broad wealth of network performance data – data that will invariably take a network expert an impossibly long time to analyze and document. |

|Furthermore, with such abundance of data, it is likely that simple problems will take a while to finally be detected, or worse still, fail to be |

|detected at all. Therefore, data mining techniques and new visualizations should be incorporated into the data collection and analysis to highlight the|

|important features and make it useful and not onerous. |

|In addition, the proliferation of data would mean that there could be tens of thousands of time series data of multiple metrics between multiple pairs |

|of hosts; and to determine the scope and locations of performance problems would be nigh on impossible. This naturally leads to automated systems that |

|will help analyze, detect, present and report on potential problems – soon after it happens or in some cases by looking at less critical network error |

|rates even predict that a more critical error is imminent as is done for disk drives [[li]]. There is therefore a need to reduce the data volume into |

|manageable and easy to understand formats (tables, graphs etc) whereby network engineers and operational managers can quickly determine the performance|

|characteristics of their network. |

|Using example techniques presented above for event detection, this proposal will also focus on the automated detection, alerting and diagnosis of such |

|network performance problems on a global scale. This will be done using heuristic models of network fault detection [[lii]] combined with the cross |

|correlation of events in time series for a given metric for a given path; also other related time series for different metrics, different target hosts |

|and probe sources; plus information from intermediate points along the network path(s) looking for shared experiences will be used to eliminate or |

|pin-point common possible causes. The ultimate goal of such complicated analysis will help increase the accuracy of problems, reduce the frequency of |

|false positives and provide more diagnostic information by eliminating possibilities and helping pin-point the possible cause(s). Alerts will be |

|generated in the form of emails with attached diagnostic information. To reduce the annoyance factor of multiple alerts we will also use techniques |

|(such as triggering an event state when one threshold is passed and resetting it when recovery passes another threshold (hysteresis)), and by not |

|un-necessarily repeating alerts for the subsequent re-detection of the same event. |

|Over the last two and a half years, SLAC has a proven track record in supervising and training Pakistani NIIT students both at NIIT and at SLAC to be |

|productive, acquire knowledge and share it and their expertise to benefit Pakistan/PERN network activities. These students have come from classes that |

|contain 50:50 ratio of male:female. In the past there have three women students who worked with MAGGIE and have recently graduated. Numerous |

|interesting and innovative projects have spawned from this relationship (previously funded under the MAGGIE project [[liii]]) which has developed |

|several measurement, analysis tools and international case studies available in the public domain. Examples include: PingER management; IEPM-BW |

|deployment at about a dozen sites; the ABwE measurement probe [[liv]]; yearly ICFA SCIC Network Monitoring Reports [[lv]] giving quantitative, |

|non-subjective measures of the Digital Divide from an Internet point of view; detailed case studies [[lvi]] of the network performance within and |

|between the region and the rest of the world for: Pakistan, South Asia, Latin America, Palestine, and Sub-Saharan Africa; and numerous prestigious |

|invited presentations around the world for example the American Physical Society [[lvii]], ESnet/Washington [[lviii]], Internet2/Washington [[lix]], |

|ICTP/Italy [[lx]], Mumbai/India [[lxi]], Pakistan [[lxii]] and ICTP/Romania [[lxiii]]; |

|SLAC has also hosted visits by the NIIT Rector, staff and faculty enabling faculty members to gain valuable experience and promote the research culture|

|in their classrooms. SLAC staff have also visited Pakistan, learnt in more detail of the needs and possibilities, and shared their expertise and case |

|studies directly with the students and faculty both of NIIT, other universities in Islamabad/Rawalpindi, and with PERN, HEC and related institutions. |

|This has resulted in useful observations and recommendations that have been made to PERN and HEC among others. |

|3(d) Scope and Objectives |

|Given the need of network monitoring and the analysis of network data in order to make it useful, the scope of this proposal is threefold: |

|To apply the perfSONAR framework to PERN, this will involve amending and or creating new services, from the existing perfSONARservices, which are |

|tailored towards the specific requirements of the Pakistani research network – i.e. the software. |

|Develop new and innovative presentation, reporting, network problem detection and alerting techniques based upon the perfSONAR framework using the |

|experience gained from previous projects and make them production quality for use on PERN2 – i.e. the analysis. |

|To foster the partnership between NIIT/Pakistan and the SLAC/U.S. to improve the quality and relevance of Pakistani higher education in science and |

|technical fields. |

|This project also aims to develop new forecasting tools and identify network anomalies as a means by which network problem detection can be extended. |

|Building on analysis techniques used at SLAC, the forecasting will use advanced statistical models such as Holt-Winters and ARMA/ARIMA. |

|To assist in divide and conquer techniques for diagnosis and pin-pointing of network performance issues, we will also work with PERN to install one or |

|more nodes in their network. A major purpose of these nodes will be to make on-demand measurements (e.g. ping, thrulay) with sufficient Authentication,|

|Authorization and Auditing. These will also assist in dividing possible problem sources up to assist in pin-pointing the cause. We will ensure the |

|software is configurable to meet different sites’ configurations. In addition we will work closely with PERN to get secure access to backbone |

|router/switch information (e.g. utilization, errors, configuration information such as location etc.). |

|Bringing PERN into the project will help the whole Pakistani educational and research community join and share experiences with a growing community of |

|worldwide NRENs. In addition, across these NRENs, there will be common monitoring and assisted diagnosis to assist users in effectively using their |

|network connections. This will be critical for users such as scientists and researchers using the NCP [[lxiv]] and NUST [[lxv]] clusters. |

|The deployment of a perfSONAR installation for PERN2 will be staged. Initial end-site installations will be at SLAC and NIIT. These will help to create|

|documentation drafts, and installation procedures while improving the deployability of the framework. Following this we shall install at end-sites |

|where we have existing strong partnerships. These include the National Centre for Physics (NCP) at Quaid e Azam University (QAU) and COMSATS where |

|there are strong scientific programs, concentrations of senior scientists and Grid activities - we believe that application at these sites will offer |

|the best use-cases for the advanced technologies and techniques outlined in the next sections. We are working to extend our partnerships to include |

|sites in Karachi (e.g. SSUET and/or FAST-NU) and elsewhere in Pakistan with similar strengths. |

|Typically the machine to host the monitoring software at a site will be provided and administered by the site with sufficient power and connectivity. |

|Using open source software and operating systems (Linux), we will reduce the cost of such machines whilst also increasing their security. Persuading |

|end-sites to contribute their own machines will also reduce the site-specific administration and security/deployment issues while improving the |

|“buy-in” partnership of the end-site personnel. |

|The expertise of SLAC’s, and more recently NIIT’s, involvement in global network monitoring will be applied to demonstrate the middleware |

|components/services of the perfSONAR framework. |

|An important part of this proposal builds on SLAC’s and NIITs continued commitment to help develop the intellect, skills and ability of Pakistani |

|students. It does this by training NIIT graduate students to contribute to research and development of computer networks to benefit the Pakistani |

|research and education community. Selected students will spend time at SLAC, which is highly regarded as one of the foremost research facilities in the|

|world; and have access to high performance computing and networking facilities not currently available in Pakistan. They will also have access to |

|advanced high-speed networks in the US and Europe, and will work with acknowledged experts in the network monitoring field. Thereby they will be |

|developing highly skilled human resources for Pakistan by conducting applied research and developing network monitoring and end-to-end network |

|performance measurement tools/infrastructure. This will greatly increase their skills in very relevant areas for Pakistan and not only will they |

|contribute to the development of new technologies but also share the experience and will be in a position to contribute positively to the Pakistani |

|society and economy. |

|As a tentative schedule, in the first year we intend to identify, extend and customize, package and deploy and validate the necessary performance |

|services across Pakistan. In the second year we intend to apply and evaluate mechanisms that will be productive in detecting and pin-pointing network |

|performance problems, of which in the third year we will ensure that the most useful are hardened into production quality services that will be used |

|across PERN and potentially worldwide. |

|3(e) Methods |

|The primary objective of this proposal is to deploy, enhance and customize perfSONAR middleware services for an Autonomous System (PERN2) in Pakistan. |

|In addition to this we intend to analyze and evaluate algorithms for network weather forecasting, event detection and diagnoses. It will deal with the |

|deployment, extension of the services and integration of new tools into the perfSONAR framework. |

|The main focus of this proposal is to utilize the expertise of SLAC’s history in network monitoring in order to help guide students at NIIT to |

|understand and contribute effectively to the global perfSONAR framework, and also towards building a useful network monitoring framework for Pakistan. |

|Collectively, a series of perfSONAR modular services will be developed and deployed to provide the performance middleware upon which new tools and |

|analysis service will monitor important devices along the PERN infrastructure and provide end-to-end performance analysis by being deployed at select |

|academic institutions around Pakistan: |

|Measurement: Measurement Point (MP) services will be developed to provide access to active and passive network monitoring tools available on PERN. They|

|will create and/or publish monitoring information related to active and passive measurements. Here, we intend to investigate suitable network |

|monitoring tools required to gather network performance statistics for analysis and determine their usefulness and feasibility. Existing tools such as |

|ping, traceroute, iperf, SNMP [[lxvi]] utilization (Router/Switch utilization), BWCTL [[lxvii]], Thrulay and OWAMP [[lxviii]] will be deployed to |

|provide the core performance metrics that will allow problem detection and analysis. |

|Storage: Measurement Archive (MA) services will be deployed to allow the storage of MP data that will provide historical information regarding the |

|performance of the network. This is an important service that will provide the necessary data to determine trends for forecasting and anomaly |

|detection. |

|Topology Service: It is a non-trivial task to engineer the network without having an understanding of the dynamic nature way in which network paths |

|diverge. Snapshots of how the complex algorithms enable the Internet to remain interconnected results in the network topology. This service will |

|provide the information regarding the topology of the network. This topology will be used extensively in visualization of the weather maps which will |

|help network engineers to analyze the network efficiently. It will also provide the data from which problems in network paths can be easily detected |

|and investigated. |

|Decentralized discovery: Maintaining the list of services manually or having static labels for them is not suitable for a dynamic environment. Thus it |

|is reasonable not only to have a lookup service, but also a decentralized lookup service to determine the access points the services. MP’s, MA’s and |

|topology services will register with. The lookup services will provide DNS [[lxix]] like mechanisms to discover the (virtual) location of the various |

|perfSONAR services. |

|Security: It is apparent that sharing information regarding Autonomous Systems with users without suitable Authentication, Authorization or even |

|Accounting (AAA) is not a reasonable concept. Thus there is a need to provide suitable AAA services to the infrastructure. This would allow for |

|authorized access to trusted user groups only with a provision for logging the events. Building upon the perfSONAR concepts, it will utilize existing |

|efforts made by Shibboleth [[lxx]] and eduGAIN [[lxxi]] to provide identity management using existing SAML [[lxxii]] standards. |

|The next major phase is to provide reporting and diagnostic services such as aggregation and summarizations of the relevant performance measurement |

|information, and to visualize it in a manner that is in accordance with the best practices. This requires that MA’s have been deployed and accessible |

|such that performance metrics can be obtained. In this regard, we intend to develop and integrate: |

|Anomaly Detection and Diagnosis Services - Currently, diagnosis of network problems requires much administrative manpower and time to determine the |

|cause of the problem(s). The need is to quickly and automatically identify and diagnose network problems and dramatically decrease the period that |

|users experience degraded performance. This aims to automatically gather relevant information, and use heuristical analysis to enable location (e.g. |

|network or host issues), identification (narrowing down of the cause), and verification (double-checking against maintenance tickets etc.) of events. |

|Alert Service - notify relevant network administrators in the event of an anomaly. Some basic functionality is already in place to provide alerts |

|within the IEPM-BW framework when performance changes occur. This project will both provide a more generic method to the automated analysis of network |

|performance data, and also investigate and implement advanced algorithms for such analysis and reporting |

|Network Performance Forecasting - The prediction of network performance will greatly aid large scale data replication applications. However, such |

|techniques are currently not very well understood. This project will investigate and report upon techniques to provide reliable forecasts of network |

|performance based upon past performance. |

|Reporting – The provision of manager level reports on the various network performances is necessary to help network managers make important choices in |

|the identification of potential network problems and divert necessary resources to those parts of the network. Similarly, high level governmental |

|reports should be provided to offer guidance on the International health of Pakistani’s Internet. |

|NIIT students under guidance from SLAC and NIIT staff will design, develop, test, integrate, install, document and support network measurement, |

|analysis and presentation tools, plus infrastructure management tools. In addition selected NIIT graduate and PhD students will spend a year at SLAC |

|working closely with SLAC’s experts in a world class research center, and take courses at Stanford University. In addition it will provide |

|opportunities for the NIIT faculty to visit the high quality research environment at SLAC and acquire the latest trends in network and related |

|technologies. Finally it will enable SLAC staff to visit Pakistan, learn in more detail of the needs, and share their expertise, and vice-versa. |

|3(f) International Cooperation |

|The division of effort between Pakistan and the U.S. is optimal. The co-supervision of NIIT students and their engagement in important network |

|monitoring projects with an international dimension adds a dramatic widening of their experience. Choosing students at NIIT to be involved in this |

|project and then further selecting just the best to spend time at SLAC is an enormous incentive. The students visiting SLAC will work with leading |

|network experts at a world class facility, experience tuition from world renowned professors and should become leaders in their field. |

|This way, not only will the student and faculty contribute to the development of new technologies but they also share their experiences and are in a |

|position to contribute positively to the Pakistani society and economy. |

|As such all funding for this proposal is aimed towards supporting student activities and contribution and is provisioned with approximately one-third /|

|two-thirds split of the stipend costs of the students. The members of this proposal in both the US and Pakistan offer their expertise and time gratis |

|as a commitment to their dedication towards this project. |

|3(g) Relevance |

|Even after embracing information technology for more than over a decade, Pakistan lacks in the availability of sufficient skilled human resource. A |

|number of reasons may be attributed to this lack of quality human resource. A major attribute to this situation may be the lack of opportunity and |

|resources. TWe believe that the the project MAGPIE addresses such issues (and achieves the funding program’s objectives) in the following manner: |

|This project will contribute to the nation’s development of human resource as the students and faculty working under this project shall be given the |

|opportunity to gain knowledge of research and development in the domain of network performance monitoring as well as valuable experiences and best |

|practices by working with renowned institutes such as SLAC. During a period of three years it is estimated that the direct beneficiaries of the project|

|will include two PhDs, two Masters, eight undergraduate students and four faculty/staff members. However the project will have a multiplier effect on |

|approximately sixteen faculty members and several research associates, undergraduate and post-graduate students as a consequence of interaction with |

|the direct beneficiaries. |

|The local industry such as Micronet Broadband Pvt. Limited [[lxxiii]] shall also benefit from the experiences since they are in close collaboration |

|with NIIT regarding three development projects (maintenance of network-route consistency, dealing with DoS attacks on the service network and open |

|software for network setup and management) to fulfill their local requirements. |

|This project will also enhance the capabilities of PERN to function as a network with over 60 universities throughout the country. The collaborators of|

|this proposal believe that jJust connecting the universities is not sufficient; planning maintenance and continual improvement of service levels is |

|also equally important, as is the ability to pro-actively respond to problems. With the team’s rich expertise in Network Monitoring and Measurement of |

|the digital divide, the analyzed data would be highly valuable to the decision makers in the government. |

|As a consequence of this project the participants will not only gain knowledge of network performance measurement and analysis, but will also have the |

|experience of working on high-performance networks. This trained human resource will then be of immense importance to diverse organizations in the |

|local industry such as Internet service providers, telecomm operators as well as the software industry. |

|3(h) Results Statement |

|Design and deploy a perfSONAR compliant infrastructure on PERN2 that will offer very valuable and varied network performance data across both the |

|backbone and selected end-sites across Pakistan. |

|Using this data, design innovative and production quality mechanisms to present, analysis, diagnose and pin-point network performance problems that |

|will be beneficial to both network engineers and Internet end users. |

|Pakistani students and faculty who participate will gain experience in high performance networks and innovative research in networks and their |

|monitoring; and be enabled to contribute to Pakistani education and growth. |

|3(i) Additional Benefits |

|Building a large international community invested in deploying the perfSONAR measurement framework is as important as developing the actual software. |

|The perfSONAR framework is being developed in an open community by consensus, with all community-developed code given a BSD-style open-source license. |

|There is thus no impediment to contributions from Pakistan or Pakistan taking benefit from the developments from this project. |

|Academic institutes participating in this project will have access to a large scale monitoring network in Pakistan which will provide them practical |

|hands-on experience. Students and researcher from these institutes shall be able to incorporate their own research ideas into this infrastructure. |

|This infrastructure will make it easier for educators to bring the network into the classroom. Early prototype components of the perfSONAR system have |

|been featured in courses at the University of Delaware [[lxxiv]]. Daily use of the network for routine teaching activities depends on a network |

|infrastructure that can meet the needs of the teaching application. A perfSONAR-enabled network will allow educators and support staff to easily verify|

|and demonstrate that the network is operating properly. |

|Members of the perfSONAR consortium are already engaged with resolving performance issues on production networks. By continuing to work with operations|

|staff, local administrators and faculty, the team will promote the tools and the advantages of self service over waiting for third parties to run |

|tests. For example Georgia Tech. and SLAC are deploying campus wide network monitoring infrastructure. These infrastructures will be perfSONAR enabled,|

|allowing faculty and central support staff to view performance and schedule tests within and beyond campus. |

|This proposal will help build bridges across the world, extend education to both male and female students, reduce misunderstanding and help reduce |

|radical, fundamentalist indoctrination of hate. |

3(j) Budget (in equivalent US $) (For Year 1)

|Items requested |Pakistan Side |US Side |

| |Requested |Approved |Requested |Approved |

|Equipment (Attach Proforma invoice) | | | | |

|Three laptop computers for students | | |$4701 | |

|SLAC Compute server for analysis and data storage | | |$4220 | |

|Compute server for analysis and data storage ($4400) - one required at |$4400 | | | |

|NIIT campus for deployment (year I and year II) | | | | |

|Consumables | | | | |

| | | | | |

|Services (Attach Proforma invoice) | | | | |

| | | | | |

|Travel | | | | |

|SLAC domestic travel (2 trips/year) | | |$3000 | |

|Three Grad students per year visit SLAC (US) air fare ($2150) |$6,450 | | | |

|Two persons (faculty, NIIT) per year visit SLAC (US) for two weeks - |$8,500 | | | |

|air fare $2150, stay per diem $150 per day | | | | |

|Two person (SLAC) per year visit NIIT, Pakistan and 2 weeks stay - air |$8,500 | | | |

|fare $2150, stay per diem $150 per day | | | | |

|International conference: Two person (faculty/research assistant) per |$6,100 | | | |

|year for one week - $2150 + $900 (air fare + DA) | | | | |

|Domestic Collaboration: One person (faculty/research assistant) per |$700 | | | |

|year for one week $700 (air fare + DA) | | | | |

|Support personnel | | | | |

|Three graduate student for 12 months ($1000/mo from SLAC plus 3.8% | | |$37368 | |

|benefits) | | | | |

|Three grad students of NIIT per year stay at SLAC(US) for 12 months |$64,800 | | | |

|(share: $1800 per month from NIIT) | | | | |

|One Project Manager ($850 per month) |$10,200 | | | |

|Stipend for six Research Associates at NIIT ($150 per month) |$10,800 | | | |

| | | | | |

|Miscellaneous | | | | |

|Indirects (15% SLAC) (3% NIIT) |$3,613.5 | |$7393 | |

|TOTAL US $ |$124,063.5 | |$56684 | |

Budget (in equivalent US $) (For Year 2, if requested)

|Items requested |Pakistan Side |US Side |

| |Requested |Approved |Requested |Approved |

|Equipment (Attach Proforma invoice) | | | | |

|Compute server for analysis and data storage ($4400) - one required at |$4,752.0 | | | |

|NIIT campus for deployment (year I and year II) | | | | |

| | | | | |

| | | | | |

|Consumables | | | | |

| | | | | |

|Services (Attach Proforma invoice) | | | | |

| | | | | |

|Travel | | | | |

|SLAC domestic travel (2 trips/year) | | |$3105 | |

|Three Grad students per year visit SLAC (US) air fare ($2150) |$6,966 | | | |

|Two persons (faculty, NIIT) per year visit SLAC (US) for two weeks - |$9,180 | | | |

|air fare $2150, stay per diem $150 per day | | | | |

|Two person (SLAC) per year visit NIIT, Pakistan and 2 weeks stay - air |$9,180 | | | |

|fare $2150, stay per diem $150 per day | | | | |

|International conference: Two person (faculty/research assistant) per |$6,588 | | | |

|year for one week - $2150 + $900 (air fare + DA) | | | | |

|Domestic Collaboration: One person (faculty/research assistant) per |$756 | | | |

|year for one week $700 (air fare + DA) | | | | |

|Support personnel | | | | |

|Three graduate student for 12 months ($1000/mo from SLAC plus 3.8% | | |$38676 | |

|benefits) | | | | |

|Three grad students of NIIT per year stay at SLAC(US) for 12 months |$69,984 | | | |

|(share: $1800 per month from NIIT) | | | | |

|One Project Manager ($850 per month) |$11,016 | | | |

|Stipend for six Research Associates at NIIT ($150 per month) |$11,664 | | | |

|Miscellaneous | | | | |

| | | | | |

|Indirects (15% SLAC) (3% NIIT) |$3,902.6 | |$6267 | |

|TOTAL US $ |$133,988.6 | |$48048 | |

Budget (in equivalent US $) (For Year 3, if requested)

|Items requested |Pakistan Side |US Side |

| |Requested |Approved |Requested |Approved |

|Equipment (Attach Proforma invoice) | | | | |

| | | | | |

|Consumables | | | | |

| | | | | |

|Services (Attach Proforma invoice) | | | | |

| | | | | |

|Travel | | | | |

|Domestic travel (2 trips/year) | | |$3210 | |

|Three Grad students per year visit SLAC (US) air fare ($2150) |$7,523.3 | | | |

|Two persons (faculty, NIIT) per year visit SLAC (US) for two weeks - |$9,914.4 | | | |

|air fare $2150, stay per diem $150 per day | | | | |

|Two person (SLAC) per year visit NIIT, Pakistan and 2 weeks stay - air |$9,914.4 | | | |

|fare $2150, stay per diem $150 per day | | | | |

|International conference: Two person (faculty/research assistant) per |$7,115.0 | | | |

|year for one week - $2150 + $900 (air fare + DA) | | | | |

|Domestic Collaboration: One person (faculty/research assistant) per |$816.5 | | | |

|year for one week $700 (air fare + DA) | | | | |

|Support personnel | | | | |

|Three graduate student for 12 months ($1000/mo from SLAC plus 3.8% | | |$40030 | |

|benefits) | | | | |

|Three grad students of NIIT per year stay at SLAC(US) for 12 months |$75,582.7 | | | |

|(share: $1800 per month from NIIT) | | | | |

|One Project Manager ($850 per month) |$11,897.3 | | | |

|Stipend for six Research Associates at NIIT ($150 per month) |$12,597.1 | | | |

|Miscellaneous | | | | |

| | | | | |

|Indirects (15% SLAC) (3% NIIT) |$4,060.8 | |6487 | |

|TOTAL US $ |$139,421.5 | |$49730 | |

Budget (in equivalent US $) (TOTAL REQUESTED for all years of project)

|Items requested |Pakistan Side |US Side |

| |Requested |Approved |Requested |Approved |

|Equipment | | | | |

|Three laptop computers for students | | |$4701 | |

|Compute server for analysis and data storage | | |$4220 | |

|Compute server for analysis and data storage ($4400) - one required at |$9,152.0 | | | |

|NIIT campus for deployment (year I and year II) | | | | |

|Consumables | | | | |

| | | | | |

|Services | | | | |

| | | | | |

|Travel | | | | |

|Domestic travel (2 trips/year) | | |$9319 | |

|Three Grad students per year visit SLAC (US) air fare ($2150) |$20,939.3 | | | |

|Two persons (faculty, NIIT) per year visit SLAC (US) for two weeks - |$27,594.4 | | | |

|air fare $2150, stay per diem $150 per day | | | | |

|Two person (SLAC) per year visit NIIT, Pakistan and 2 weeks stay - air |$27,594.4 | | | |

|fare $2150, stay per diem $150 per day | | | | |

|International conference: Two person (faculty/research assistant) per |$19,803 | | | |

|year for one week - $2150 + $900 (air fare + DA) | | | | |

|Domestic Collaboration: One person (faculty/research assistant) per |$2,272.5 | | | |

|year for one week $700 (air fare + DA) | | | | |

|Support personnel | | | | |

|Three graduate students for 12 months each year for 3 years | | |$116073 | |

|Three grad students of NIIT per year stay at SLAC(US) for 12 months |$210,366.7 | | | |

|(share: $1800 per month from NIIT) | | | | |

|One Project Manager ($850 per month) |$33,113.3 | | | |

|Stipend for six Research Associates at NIIT ($150 per month) |$35,061.1 | | | |

|Miscellaneous | | | | |

| | | | | |

|Indirects (15% SLAC) (3% NIIT) |$11,576.9 | |$20147 | |

|TOTAL US $ |$397,473.6 | |$154460 | |

OTHER FUNDING. Please list the source and amount of any other funds received or applied for to support this project:

3(k) Executive Matters / Budget Requirements

|i. Timetable (Please indicate each major step in project evolution referring to the time schedule) |

|[pic] |

|Figure 5 Time line of project tasks |

|ii. Request justification (Please provide detailed justification for each budget item requested) |

|Support Personnel: |

|The nine graduate students who will be visiting SLAC for one year each, will be funded from the SLAC side at $1000 per month. The remaining support of |

|$1800 as peris based on the requirement that SLAC must pay twice California minimum wage [[lxxv]] for exempt graduate student researchers. This willshall |

|be provided by NIITUST Institute of Information Technology, Pakistan. |

|NIIT shall employ a full-time project manager to ensure the accomplishments of tasks within deadlines. His job description shall not only include planning|

|(i.e. with support from the research staff) and scheduling of tasks but also the documentation of the proposed project’s progress and liaison with the |

|funding agency, in this case HEC. The research staff shall focus on the implementation of tasks, software development and testing, research and analysis |

|etc. |

|Research Associates (RA) shall also be employed at NIIT to support the research and development activities. These RAs may be graduate students as well as |

|undergraduate students conducting their thesis projects (or utilizing their spare time from studies) in development activities under the umbrella of this |

|project – MAGPIE. |

|Miscellaneous Expenses: |

|Over the three years SLAC will need a laptop for each of three students currently at SLAC. This will enable them to work when traveling, at home, |

|telecommuting etc, Since laptops have a rough lifetime of three years we are asking to procure three laptops in the first year and none thereafter. SLAC |

|will provide system installation, major software applications and maintenance at no cost to the project. We estimate the cost of each laptop to be $1567. |

|Also in the first year SLAC will procure a rack mountable server and disk space to host the data and analysis tasks at a cost of about $4220. The |

|administration, network connections, power, maintenance etc. will be provided by SLAC at no cost to the project. |

|All SLAC costs will be escalated at 3.5% for years two and three. |

|The Stanford University fringe benefit rate of 3.8% for graduate students has been applied to salary costs, |

|SLAC’s indirect costs will be charged at 15% to all items, following the proposal instructions, although SLAC’s normal indirect rate is 39%. Dr. Les |

|Cottrell, head of the Network Computing Group, will direct the project at SLAC (10% effort / no cost to the project). Other researchers (35% FTE/no cost |

|to project) will provide coordination, guidance, training, quality control and integration. |

|SLAC is a department of Stanford University, under a Management and Operating contract with the US Department of Energy. All SLAC expenditures under this |

|grant are incurred under the cost principles governing SLAC as outlined in the DoE/SLAC contract. |

|NIIT shall utilize its existing resources available in the research labs and provide desktop machines to the research and development teams. Relevant |

|software applications and maintenance shall also be provided with support from the University funds. |

|However, NIIT shall require two rack mountable servers, one for production software and the other for development and testing. The software shall be |

|deployed in the form of server applications such as Measurement Points, Measurement Archives, Topology Discovery Services and miscellaneous software. NIIT|

|plans to deploy these software initially at NIIT’s premises and later (in various stages) at campuses of major collaborating partners such as National |

|Center for Physics (NCP), Quaid-e-Azam University, PERN and COMSATS Institute of Information Technology (CIIT). |

|iii. Travel details (Please indicate number, duration, timing, and justification for each visit proposed for each side) |

|Research Students visit / stay at SLAC (US): A major part of this collaborative effort is to provide students with high quality research environment. Nine|

|students are to be sent to SLAC over the three year duration for research visits of one year’s duration for each student. During this visit students will |

|have a chance to grow in an environment which has a proven track record of basic and applied research. During this visit students will also be able to |

|take courses at Stanford University. |

|NIIT Faculty visit/stay at SLAC for collaborative meetings: This collaborative project requires frequent meetings between NIIT and SLAC mentors and |

|students. Most of the communication is through e-mails, phone calls and video conferences which are very useful but not enough. It is required that |

|faculty of NIIT working on this project visit SLAC for a period of 2 weeks each year so that faculty members are able to guide students in a better |

|fashion. |

|SLAC supervisors visit to NIIT: Some students working with this research group will have a chance to visit SLAC as internees. For guidance of other |

|students it is required that a SLAC mentor visits NIIT once a year for a period of two weeks each year, so that even if a student does not visit SLAC |

|there is a strong interaction with existing students and new students can be attracted and introduced. |

|Conference Travel & per diem: We are anticipating two publications per year in this project. We have decided to publish one of those in international |

|conferences to establish the usefulness and importance of our research. We have also decided to publish one of the research papers in a domestic |

|conference so that other Pakistani institutes and students also benefit from our work. |

|Local Pakistan travel: This category involves travel by Network Assistant or research student for the installation/upgrade/ deployment of the server(s). |

|This is an ongoing activity which shall continue for the whole duration of the project. |

|SLAC requests funding to cover the cost of two domestic (U.S.) trips/year to attend conferences, present findings, meet with and share information with |

|collaborators in perfSONAR. |

|iv. Additional notes (Please provide additional appropriate information, if any) |

| |

3(l) Curricula Vitae (Please attach brief CVs—1 or 2 pages each—for all Pakistani and US participants. Please do not include lengthy publications lists or copies of publications.)

Co-PI (Pakistan):

Prof. Dr Arshad Ali is a senior IT professional and academicians. He completed his PhD studies in 1992 from University of Pittsburgh, USA in the Design, Simulation and Fabrication of Optical Waveguides for Optical Computers. He has over 26 national and international publications to his credit. Dr Arshad has been awarded Pakistan Academy of Sciences and COMSTECH gold medal for IT research. Currently he is working as Dean, NUST Institute of Information Technology (NIIT), Rawalpindi.

Dr Arshad enjoys good rapport with IT industry which helped in establishment of CISCO Local Academy program, attracted donation of state of the art computer Lab from INTEL, USA and a teradata Lab from NCR Pakistan for NUST Institute of IT (NIIT).

Dr Arshad initiated research collaboration with Center for European Nuclear Research (CERN), Switzerland and earned the Associate Institute status of CMS-CERN for National University of Sciences & Technology (NUST). Dr Harvey Newman from CALTECH USA (29 faculty members include Nobel Laureates) has committed 40,000 US dollars per year grant to strengthen this research collaboration initiative. The research collaboration with Stanford University (SLAC Project), USA has also helped in attracting 100,000 US dollars funding from US State department.

Dr Arshad formed a joint consortium with University of the West of England UK, Beijing Institute of Technology China, University of Savoie France and NUST Pakistan which attracted 400,000 Euro research funding from European Union. He is actively pursuing research in Distributed computing and Grid Technologies in collaboration with University of Melbourne Australia and University of Portsmouth UK. Prof. Dr Arshad is also contributing in launching of open source resource center in collaboration with Pakistan Software Export Board, Comtec (Communication Technologies) Japan and NUST, Pakistan.

Co-PI (US):

Dr. R. Les. Cottrell received his BSc (Hons) in 1962 and his PhD in Nuclear Physics in 1967 from Manchester University, England. He then joined the Stanford Linear Accelerator Center (SLAC) as a research physicist in High Energy Physics, focusing on real-time data-acquisition and analysis in the Nobel 1990 prize winning group that discovered the quark. In 1972/3 he spent a year’s leave of absence as a visiting scientist at CERN in Geneva, Switzerland. In 1979/80 he spent another year’s leave of absence this time as a visiting scientist at the IBM UK Laboratories at Hursley, England. While at IBM he developed a dynamic graphical cursor that was granted United States Patent 4,688,181.

On returning from IBM UK, in 1980, he became the manager of SLAC’s computer networking. In 1982 he became the Assistant Director of SLAC’s Computer Services (SCS). From 1995-1997 he was the Acting Director of SCS. He is currently the Assistant Director of SCS and leads SLAC’s computer networking, telecommunications, and remote collaboration support groups. He is also SLAC’s representative on the ESnet Site Coordinating Committee (ESCC), the chairman of the International Committee on Future Accelerators (ICFA) Standing Committee on Inter-regional Connectivity (SCIC) Network Monitoring Working Group. He has served on many advisory groups such as for the Superconducting Super Collider (SSC), the American Institute of Physics, IHEP/Beijing, Internet2, and FNAL, as well as technical committees such as chairing the ESCC Network Monitoring Working Group, and being a member of the Global Grid Foundation’s Network Monitoring Working Group. He is also the PI of the DoE funded Internet End-to-end Performance Monitoring (IEPM) project which has attracted funding of almost $2M since 1997.

He was the leader of the effort that in 1994 established the first permanent Internet connection to mainland China. He was co-PI of teams that won the Internet2 Land Speed record twice, an achievement listed in the Guinness Book of Records (2004). He was a co-leader of teams from Caltech and SLAC and other institutes that won the prestigious SuperComputing Bandwidth Challenge three years in succession. He is the SLAC PI of the DoE funded Terapaths project, the NIIT/MAGGIE project, and the collaboration with ICTP, Trieste.

4. References

-----------------------

[i] M. Desai, S. Fukuda-Parr, C. Johansson, F. Sagasti, "Measuring the Technology Achievement of Nations and the Capacity to Participate in the Network Age", Journal of Human Development, Vol. 3, No. 1

[ii] H. Newman, “ICFA SCIC Report for High Energy Physics” available at

[iii] “Network Problem Cases”, available at

[iv] perfSONAR, available at

[v] LBL, “tcpdump”, available at .

[vi] NLANR, “OCXMON”, available at .

[vii] Cisco, “Netflow”, available at .

[viii] “Web100”, available at .

[ix] NLANR, “Iperf”, available at ,

[x] Internet2 “Thrulay”, available at .

[xi] Internet2, “Network Diagnostic Tool”, available at

[xii] Internet2, “Network Path and Application Diagnosis”, available at .

[xiii] NLANR, “Introduction to the NLANR AMP Project for HPC sites”, available at

[xiv] Advanced Network Services, “Surveyor”, available at .

[xv] IEPM: “Internet End-to-end Performance Monitoring - Bandwidth to the World (IEPM-BW) project”, available at .

[xvi] W. Matthews and R. L. Cottrell “The PingER Project: Active Internet Performance Monitoring for the HENP Community”, IEEE Communications Magazine, May 2000.

[xvii] RIPE, “Test Traffic Measurement Service”, available at

[xviii] HEC, "Request for Proposal (RFP) for Establishment of 10GbE Metro Area Network for PERN2 in Islamabad/Rawalpindi Region", Higher Education Commission, H-9 Islamabad, document HEC/RFP/PERN2/564-2006, available at

[xix] Aga Khan University, “Clinical Epidemiology Unit”, available at

[xx] Pakistan Medical Research Council, “Research Activities”, available at .

[xxi] CERN, “Pakistani Physics”, available at .

[xxii] National University of Computer and Emerging Studies, “2nd Workshop on Agro-Informatics”, available at

[xxiii] Associated Press of Pakistan, “Media University in Pakistan – a significant leap into the future”, available at

[xxiv] National College of Arts, Lahore, “Research and Publication Centre”, available at .

[xxv] “Virtual University of Pakistan”, available at ..

[xxvi] SLAC, “Throughput from SLAC to Caltech reduced by a factor of 5”, available at

[xxvii] See for example: Wachovia, “Wachovia Benefits from Improved Inter-Office Collaboration with Cisco TelePresence”, available at .

[xxviii] CERN, “Large Hadron Collider”, available at .

[xxix] NIIT/Cisco “Health-e-Pakistan - Developing a Pakistani Health Grid Network”, proposal in progress, contact Dr. Amir Shafi ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download