F5 Data Manager Sample Report and Analysis

[Pages:16]F5 Data Manager Sample Report and Analysis

F5 Data Manager provides extensive reporting capabilities that can help IT organizations better understand the profile and characteristics of their file data. It highlights opportunities to optimize and improve file data management strategy over time.

Contents

Introduction

3

Client Case Study: Background

3

Detailed Reporting

3

Total Storage Capacity

4

File Age

5

File Type

6

Top Owners

7

Trend Reporting

8

Storage Usage

8

Top Owners

9

Advanced Reporting

10

Top Owners

10

Return on Investment

11

Analysis

11

Statistics in Data Manager Reports

12

Data Manager Trial

13

2

Introduction

Enterprise organizations are struggling to keep up with the rapid growth of the file data they must manage. With IT groups stretched to the breaking point, it often seems like the easiest solution is to continue purchasing more storage capacity. However, this can lead to unsustainable storage and operational costs. To effectively reduce these costs, organizations must better understand their file storage requirements. F5? Data ManagerTM provides a lightweight reporting solution that gives organizations visibility into their file data and file storage requirements. This paper will discuss how Data Manager reports can be used to visualize how data is growing. It includes charts and tables that are excerpts from an actual Data Manager report, plus brief analyses of the data. The analyses were provided by an F5 Networks expert to the customer to help them understand what Data Manager had uncovered and what corrective action, if any, was required. All identifiable customer information has been removed.

Client Case Study: Background

This company (whose name is being withheld for privacy) is a fast-growing, mid-size enterprise. Their IT organization was struggling to cope with storage growth rates that exceeded 50 percent annually, which was causing a series of problems: ? Individual users and applications were frequently hitting the capacity limits on their network shares, even though

the aggregate storage utilization for the entire organization was below 50 percent. ? Provisioning additional capacity and moving data required constant IT attention and often required client and

application downtime. ? Backup times were growing increasingly lengthy and often exceeding available backup windows. Some full

backups were taking 36 hours to complete. ? Data migrations due to consolidations and upgrades were painful and time-consuming and often required

business downtime. ? Costs for managing the explosive data growth were increasing unsustainably, far outstripping growth in their IT

budget. This company was considering implementing a file virtualization solution to resolve these issues. However, the IT organization wanted to first understand its existing file storage environment and identify particular areas of need. This company used F5 Data Manager, a software solution that is installed on a server running Microsoft Windows, to inventory its file system. Data Manager provided reporting and capacity forecasting capabilities, then helped the IT team identify key trends and problem areas that could be improved with file virtualization.

Detailed Reporting

Data Manager presents a wealth of information about file data, file systems, and file storage devices. The standard detailed reports it generates are useful for visualizing a file storage environment at a specific point in time. A good strategy for understanding this information is to examine characteristics of the overall environment, identify potential areas of concern, and then investigate those specific areas.

3

Total Storage Capacity

Figure 1: Disk capacity statistics as displayed by Data Manager Analysis The company used Data Manager to perform an inventory of all file shares in its environment--a mix of NetApp FAS, EMC Celerra, and Windows file servers. The cumulative storage capacity report showed that, despite the frequency of individual users and applications running out of capacity in their file shares, their aggregate level of storage utilization remained low. The company was spending additional money purchasing new storage every year, even though it had significant stranded capacity elsewhere in its environment. Action The company decided to virtualize its file storage environment and implement a global namespace with dynamic capacity balancing policies. This would enable IT to: ? Create a unified storage pool and automated policies to distribute storage utilization across many systems. ? Prevent localized spikes in storage utilization and even out the aggregate utilization across the entire environment. ? Set higher target levels of storage utilization without worrying about individual users and applications hitting

capacity limits. ? Redeploy existing unused capacity to where it's most needed.

4

File Age

Figure 2: Breakdown of file data by age as displayed by Data Manager Analysis Next the company looked at the age breakdown of its file data and discovered that almost 75 percent of all files had not been modified in over three months, and over 40 percent had not been modified in six months or more. Inactive file data was consuming the company's performance-optimized primary storage and driving the need to purchase additional performance-optimized capacity. In addition, the company was having issues making its backup windows, as weekly full backups were often taking over 36 hours to complete. Backups were a major source of operational pain, and by running Data Manager, the company's IT group was able to assess the impact of removing 80 percent of its inactive file data out of the weekly full backups. Action The company decided to move to a tiered storage environment. However, because this was a major change, it also planned to start with a conservative tiering policy and increase the aggressiveness over time. It decided to implement two storage tiers, augmenting existing primary storage as Tier 1 with capacity-optimized deduplicated SATA storage as Tier 2. Once the environment was virtualized, the company could create automated storage tiering policies based on the file age to: ? Automatically move data to more cost-effective storage without affecting user access to the data. Initially, files

older than 180 days will be moved to Tier 2. Once the company becomes more comfortable with storage tiering, it is considering moving files older than 90 days to Tier 2, and files older 180 days to a new Tier 3. This policy would also allow them to consider cloud storage for long-term archive. ? Reduce backup time and backup costs by establishing separate backup practices for each storage tier. By backing up Tier 2 every three months, the company reduced weekly backup times and costs by 66 percent, from 36 hours to under 12 hours to complete. It also reduced backup media costs by 70 percent by reducing the size of weekly full backups.

5

File Type

Figure 3: Breakdown of file data by extension as displayed by Data Manager Analysis Next, the company wanted to gain better visibility into what types of data users were generating by looking into user home directories, and specifically their top file extensions by capacity. As part of this process, the company discovered that email archive (.PST) files were consuming over 22 percent of the total storage capacity in its home directories. Action The company decided to augment its storage tiering implementation with additional tiering policies based on file type. Specifically, it would automatically place all .PST files on a separate storage tier comprised of capacity-optimized deduplicated SATA storage. By doing so, the company would be able to: ? Reduce backup time and backup media costs by establishing specific backup practices for .PST files. By customizing

the backup policy for this data set, the company reduced weekly full backup times and cost by over 20 percent. ? Reduce the cost of storage for a type of file that is infrequently accessed or modified. ? Maximize the benefits of data deduplication. Because email archive files belonging to multiple users often contain

many of the same attachments, data deduplication would further reduce the amount of storage capacity required for storing .PST files.

6

Top Owners

Figure 4: Breakdown of file data by owners as displayed by Data Manager Analysis The company also wanted to see how much storage capacity their top users were consuming. What they discovered was that the top ten users (out of 753) were consuming almost a terabyte of storage, or 7.6 percent of total capacity used in the company's home directories. In addition, many of the top users had their home directories on the same file servers, which was causing utilization rates to skyrocket, requiring frequent IT intervention to ensure enough space was always available for all users. Action The company was already planning to implement dynamic capacity balancing policies in its virtualized storage environment that would distribute data across all of their file servers and even out the aggregate utilization. IT also decided to more proactively monitor the file storage requirement trending for their top users.

7

Trend Reporting

Data Manager enables organizations to report on various aspects of file data growth through the creation of trend reports. A trend report illustrates how different aspects of the storage environment continue to change over time.

Storage Usage

Figure 5: Changes in storage usage over time as displayed by Data Manager Analysis The company took a closer look at one of the file shares that contained home directories for several of their top users. They discovered that one in particular was rapidly nearing its capacity limit and would soon affect users in that share. Action Although the company was planning to virtualize its environment and create dynamic capacity balancing policies, IT was still interested in the root causes of the current rapid utilization spike and planned to further investigate the storage consumption of individual users in that share.

8

 ................
................

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

Google Online Preview   Download