The next BriefingsDirect case study discussion targets how biotechnology services provider Acorda Therapeutics has implemented a strategic disaster recovery (DR) capability to protect its highly virtualized IT operations and data.
See how Acorda Therapeutics’ use of advanced backup and DR best practices and products has helped it to manage rapid growth, cut energy costs, and gain the means to recover and manage applications and data faster. Also learn how these advanced DR benefits have led to other data center flexibly and even migration benefits.
Sharing more detail on how modernizing DR has helped improve many aspects of Acorda Therapeutics’ responsiveness is Josh Bauer, Senior Manager of Network Operations at Acorda Therapeutics in Hawthorne, NY. The discussion was moderated by Dana Gardner, Principal Analyst at Interarbor Solutions. [Disclosure: VMware is a sponsor of BriefingsDirect podcasts.]
Here are some excerpts:
Gardner: What do you perceive as being different today about DR than just a few years ago? Is this really a fast-moving area?
Bauer: One of the most prominent changes is recovery time. You no longer need to restore from physical tape and see recovery times of upwards of 24 hours, something that we hadn’t seen until recently. We implemented Site Recovery Manager (SRM) from VMware and we can now do that same recovery in about four hours.
We're constantly replicating using RecoverPoint and we can get data up to the minute, versus tape, where you are at the whim of whether the backup completed on time -- did everything go to tape, and when was it done? It could have been two days ago, versus now, when it's data that’s 100 percent synced up to a minute ago.
When we had about 80 employees, we probably barely had a terabyte, and now with 350 employees we easily have over 14 terabytes.
Gardner: I am also wondering, because you are in the healthcare and biotechnology field, are there aspects of the new DR that appeal to you from a compliance or regulatory perspective as well?
Bauer: Definitely. Four times per year we have to prove that we can recover all of our software and data by doing a DR test. Until we had SRM, we had to do it all from tape, from a cold facility, and it would take us a day, sometimes a day-and-a-half. That’s just not the best way to do things. But now, with SRM, we can always do these tests on the fly, even from our office, from home, or from wherever.
Gardner: Tell me a little bit more about Acorda Therapeutics. You were founded in 1995. Tell us what you do, so our audience can understand the type of company you are and type of products and services you provide.
Bauer: We create treatments for people with multiple sclerosis, spinal cord injuries, or other neurological disorders. We have two marketed drugs in the market right now, the most recent of which, Ampyra, helps people with multiple sclerosis walk better, and it has been a huge success. And that's the main reason we've been growing so much lately.
Prior to virtualization, we were spending a lot of time managing our infrastructure, with all those physical servers. Once we virtualized everything, we spent way less time managing the infrastructure and could spend more time helping the business.
In fact, the IT department itself has become less like a computer repair shop and more like a strategy center. I'm constantly being brought into projects to help the business make the right decisions when it comes to any type of technology.
The next logical step would be to have my team spend less time doing these four-times-a-year DR drills the way I described before. With SRM it’s a few clicks. We're saving so much time and we are able to do other things.
Gardner: Tell me how you got to the point today, where you can deal with something like 14 terabytes and moment-by-moment backup capability?
Bauer: It all really started at VMworld. That’s been a fantastic way for me to learn what's out there, what's coming up, and just staying in the know. That’s actually where I met International Computerware, Inc. (ICI), who is one of our strategic partners for storage and virtualization.
I had approached them with the growth issue. We had already started doing virtualization on our own. I had used it at a previous company, but I wasn’t familiar with SRM, and it looked like it might be a nice fit for improving our DR. So ICI came in and they sort of held our hands and helped us with that project.
Specific to storage, they have also helped us make sure that we do better management of growth, anticipate our growth, and show that we have more than what we're going to need, before the growth happens, and they've done some analysis on like what we have. We brought them in before things got too bad.
Since using VMware, we've noticed uptime upwards of three nines monthly. Before that, when we were mostly a physical environment, it was nowhere near that much. We had physical servers going down all the time.
VMware immediately gained our trust, seeing that they came out with this product for DR. It was a name that we trusted. Then, we played with it for a while, and it worked out fantastically.
It's all about trusting VMware and then, again, ICI, working with them. They just know their stuff. We have a lot of different partners we work with, but we prefer to use ICI, because they really focus on doing things properly. It's more about working with someone that really knows what they are doing. They understand that we have some skills, as well. They're not trying to sell us something we don’t need.
95 percent virtualized
We are 95 percent virtualized here. The only thing that’s not virtual is our fax server, which requires a physical fax board and that’s about it. Everything else is virtual.
Gardner: So this is across all tiered apps, tier one, three, four?
Gardner: I understand you're using vSphere 5. You're on vCenter SRM 5. That only came out towards the end of last year. So you just jumped right on that.
Bauer: Oh, I didn’t waste any time. We were very excited about it, especially this new option of using a failback, which wasn’t really part of SRM Version 4.
If you ever have the very unlikely event of a a disaster, when you do a recovery, you're now operating off of the disaster equipment or recovery equipment. While that’s happening, people are still saving files and generating new data. If you were to just simply turn on the original equipment again, all that data would be lost. So you need to fail back to re-sync everything.
With SRM Version 4, you had to configure two one-way recovery systems. So it would take a lot more time. But now with failback, it's a lot more smooth, kind of built-in.
Gardner: Do you actually have separate data centers that you are backing up to? What's the topology or architecture that you're using?
Bauer: We have two separate data centers, recovery and production. At the moment they're only a few towns apart, but we are shopping around for a data center much further away. We hope to do that in the next six months or so.
Gardner: Looking to the future, one other area I wanted to hit on, which is important to a lot of folks, especially in some overseas markets, is this issue about energy. Did you have any impact on energy and/or storage costs associated with the total life cycle of the data?
Bauer: We reduced the footprint by easily 75 percent by not needing so many physical servers. That’s a pretty huge shout-out to VMware there. Also, we're not using that much power. We don’t need as big a data center. Not as much cooling is needed. There's a whole assortment of things, when you take out all the physical servers.
Gardner: Now, looking to the future, other areas that people have described as a segue from going to high virtualization, exploiting the latest technologies in DR, is to start thinking about desktop virtualization infrastructure (VDI) and desktop-as-a-service. They're even looking at cloud and hybrid-cloud models for hosting apps, then backing them up and recovering them in different data centers, which you've alluded to. Do you have any thoughts about where this could possibly lead?
Bauer: In fact, if you were going to ask me what my next initiative was going to be, and you didn’t mention desktops, that’s the first thing that would have come to mind. We're starting to explore replacing our laptops with virtual desktops. I'm hoping this is something that we could implement next year.
Right way to go
This seems like the right way to go, because our helpdesk team spends too much time swapping out laptops or replacing laptops that are dropped on the ground. You're looking at a small thin client, which is the fraction of the cost of a laptop. Plus, the data is no longer kept in a laptop. There are no security or compliance issues. You can l just give them a thin client, and they are back in business.
It makes everybody in this company, especially at the top-level, nervous to know that some sensitive data still does make it out to the laptops. We tell people to save everything to their network drives, but without using thin clients and virtual desktops, there's no other way to force that.
Gardner: How about advice for those folks that might be moving towards a more modern DR journey, as you described it? What would you advise to them as they begin, and what lessons might you have learned that you could share?
Bauer: First off, do it. You're going to be glad that you did. The good thing about this is that you can do it in parallel with your current DR plans. You don’t have to change your existing recovery plans. You can take as much time as you want to set it up right. And the key is to set up a demonstration for the key business owners and players that are going to make the decision on the change.
Set it up right with a handful of important apps, important VMs, and then just show it to people. Once they see how great it works, you're definitely going to want to change.
It's always helpful to have some outside help. No matter how skilled you are, it's always good to have a second pair of eyes look at the work that you did, if for nothing more than to confirm that you've done everything you could and your plans are solid. It's helpful to have a partner like ICI.
- Case Study: Strategic Approach to Disaster Recovery and Data Lifecycle Management Pays Off for Australia's SAI Global
- Virtualization Simplifies Disaster Recovery for Insurance Broker Myron Steves While Delivering Efficiency and Agility Gains Too
- SAP Runs VMware to Provision Virtual Machines to Support Complex Training Courses
- Case Study: How SEGA Europe Uses VMware to Standardize Cloud Environment for Globally Distributed Game Development
- Germany's Largest Travel Agency Starts a Virtual Journey to Get Branch Office IT Under Control
- Virtualized Desktops Spur Use of 'Bring You Own Device' in Schools, Allowing Always-On Access to Education Resources