Each and every month, we conduct about 50 assessments with our Ninja-Protected Backup Analyser, a software tool that collects information about a customer’s backup processes. The assessments, which are part of our BURA (Backup, Recovery and Archive) solutions program, are essentially a health check on the customer’s storage environment. Just like people need regular health check-ups, your backups need check-ups, too, so you know what’s really going on.
The assessments have become one of our most popular services over the past few years. The process takes from five to 40 minutes, depending on the size of the environment. To be sure, none of the data we collect is personal. We just gather system process information and provide concrete feedback, but here’s the bonus: We do it free of charge.
Each assessment determines things such as the number of backup policies you have in place and how frequently backups run. It determines metrics such as backup duration and maximum backup window, as well as data on backup failure ratio and, conversely, backup success ratio. The assessment also gives you data on how many restores are performed and their success rates.
The backup report is where it all comes to life
After we analyze the data, you receive a report (also free), which gives you a view into how well your backup servers perform. This also provides a benchmark for any future assessments. To date, we’ve done more than 1400 assessments representing 846 PB of data from over eight million backup jobs across various sectors and industries. Here comes the fun part of this blog: I’d like to share some of the trends we’ve detected.
Backup windows are a concern
Backup windows are “a concern” for many organizations we survey. More than half the customers assessed had backups that took more than 96 hours to complete. That’s more than four days! What’s being done with the data that continues to pour in over that four-day window? What if there is a failure three-and-a-half days in? Those long windows can be dangerous to the success of your backups. It is eye-opening, actually. The assessments also show that about 10% of customer backups take more than 192 hours. That’s more than a week! As a general rule, if it takes more than a weekend for a full backup (or longer than overnight for an incremental backup), you’ve got problems that need to be addressed.
High backup failure rates
Almost 37% of all backup jobs that we assessed had success rates of 95% or less. This means that more than five percent of backup jobs fail at some time and only 39% had better than a 98% success rate. These numbers are quite surprising and perhaps beg the question, “Do you know when your backup jobs are successful or not?”
Backup architecture is complicated
We partly attribute high failure rates to the fact that traditional backup architecture is complicated to manage. The fact that many organizations currently experience a generational shift in the data center compounds that challenge. For example, if the employee who has managed backups for the past 20 years retires and their replacement is not used to dealing with five-year-old backup technologies, there will be problems. Often the high failure rate can be traced to the sheer complexity of running backups in large environments.
Some store their data ‘forever’
Our assessment allows organizations to see exactly how much data they keep and, more importantly, for how long. We’ve met with a lot of customers who hold on to data for unlimited periods. In fact, more than 40% keep their data 366 days or longer. What does this mean? It means a big chunk of the organizations assessed back-up data they’ll never use. We ask a lot of customers why: Is it a business requirement? Did it happen by chance? Many say they hang on to the data just in case. “I might need it in the future, so I don’t delete it just in case I get sued, or in case I might need to restore it.” It’s amazing when you think about hanging on to petabytes of data that will likely never be touched again.
File servers are the problem-children of the data center
22% of all backups involve file servers, which tend to be the problem-children of the data center in terms of the time required to perform a backup. Other file systems such as Oracle, SAP and Exchange systems can also take a long time. Still, the primary culprit for crawling backups is those pesky file servers. They can take days or even over a week to back-up, based on what we’ve seen in the assessments.
The results can be shocking
We’ve worked with many customers who are shocked by what we uncover in their assessment. At first, they want to refute the findings, but once they investigate on their own, they see confirmation of what’s been uncovered. Then they begin finding ways to improve the situation. Many realize they have backup issues, but most never put the backup issues on the priority list. With our assessment results report in-hand, you can go up the chain to senior management and do something about the findings before disaster strikes. Think of it as an independent evaluation that shows you how to get up to current IT standards.
Curious about the health of your own backup system? We can perform an assessment for you, too. Get important, environment-specific information about deduplication opportunities and benefits, uncover issues around backup windows and failed backups, and find opportunities for cost reductions, such as how to safely reduce data-retention times. You’ll also receive relevant insights into your current backup environment, such as daily data backed-up, throughput data, total capacity used for backup retention, top 10 backup jobs, and data types being backed-up.
You can click here to learn more about our backup assessment analyzer and what it can do for you.
Click on this link to learn more about HPE 3PAR StoreOnce Backup solutions.