These days I spend a fair amount of my time working with IBM i clients of all sizes to enhance their backup strategy. Or in some cases just crafting a solid backup strategy that they could actually recover from. Recoverability remains the most important driver for point in time backups. Unfortunately, due to a lack of knowledge, demands for uptime, outdated hardware, or just simply a lack of care for business resiliency, some organizations fall into a reality where they have little faith in their recoverability.
I cannot stress how important your backup strategy is and will be, as we operate in a world of ransomware and cyber-attacks.
We have noticed a major uptick in IBM i shops that have business operations taken down due to cyber-attacks. I am going to try not to get on my Security soap box here and stick with point in time backup strategy. So I will keep the security talk brief. One of the best things you can do to protect your organization from a security threat, is have a solid backup. Sure, there are many things we can do to position ourselves well for preventing an attack. Having a solid backup gives you the piece of mind, that you can get back up on your feet if you ever get knocked down.
Great news is we can greatly enhance the quality and drastically reduce the impact point in time backups have on the business and its users with IBM’s FlashSystem external storage portfolio. Many IBM i shops continue to shy away from external storage due to their perception of cost and the fear of additional complexity. While I can’t argue that external storage becomes another thing to manage when you are used to internal disk and IBM i just being a rockstar at managing it, I think you would be doing yourself a disservice if you didn’t at least look at the value it can serve up to the business before dismissing it.
One of the biggest reasons IBM i shops shortchange their backups is due to demands for uptime.
Backing up to LTO tape or even a VTL requires downtime and for shops running 24×7 operations, this becomes a major problem. With IBM FlashSystem external SAN storage and a tool called FlashCopy, which is built into the software stack called Spectrum Virtualize, you are able to get a full system save with a very short pause for end users. Imagine being able to complete a full system save to tape with a short 15 to 45 second pause for end users.
Sounds almost to good to be true right? Here’s how we make it happen.
When the Flashcopy is initiated through the IBM Lab Services Full System Flashcopy Toolkit, it first pushes everything in memory down to the storage. Once it has everything down to storage, it puts a brief hold on user traffic and takes a snapshot (Flashcopy) of those source volumes on a separate IBM i LPAR, which acts as the FSFC landing zone. The data isn’t actually copied to that LPAR. A bitmap is created on both the source and target which maintains point in time integrity by hard writing only the things that have been changed on the source to the target Flashcopy LPAR. This consistency is maintained for the life of that individual Flashcopy. So for a backup, it is monitoring and hard writing changes for however long it takes to backup that partition. Once the backup is completed the FSFC Toolkits runs a BRMS report (If you use BRMS) as if it were the source LPAR, then ends the Flashcopy relationship, and destroy the environment. It then just waits for the next FSFC to be kicked off and goes through the whole process again.
If you are in a position where you can’t, or prefer not to impact your end users at all, completing a FSFC on your target HA partition is also an option. By utilizing Flashcopy for the backups, you are able to pause replication briefly, execute your FSFC, and then release replication back to keep source and target environments in sync. Add in a VTL on both sides with replication and you have yourself a top-notch backup solution.
Flashcopies do not have to be limited to just backups. Many IBM i shops use them for creating a test or development environment, reporting, or building out of training partitions.
When it comes to IBM i external storage can fill some big gaps and serve up tremendous value for your IBM i stack. Large outage windows for IBM i need to be a thing of the past. This can be the difference between leadership looking at IBM i as an old, outdated, legacy platform or a strategic platform the business can grow with. The gap between the cost of internal and external storage may also surprise you.
Have questions or want to talk about config options and pricing? Get ahold of me and we can see what IBM FlashSystems can do for you and your business.
More from this month:
- Deploying Access Client Solutions Centrally with Java Included
- Economic Cost of Falling Behind
- How Do We Fix The Great IBM i Resource Shortage?
- iTech iTip Videos
- Webinar: The Value of Staying Current
- Sips & Tricks: Coffee with iTech
- iBasics: IBM i Education for the Beginner System Administrator
- Let iTech Take You Out to the Ballgame ⚾
- Upcoming Events
- iTech Spotlight
- IBM i, FSP, and HMC release levels and PTFs (June 2022)