Ask an MSP Expert

Q: We’re a VAR, and we’re moving to more of a managed service model. We started out slowly by adding a few managed services to our offering, and we recently added backup and recovery. Of course we hope we won’t have to restore a customer’s data anytime soon, but we know it’s inevitable. What are some tips and best practices we can use to make sure we have fast, efficient restore times?

You’re certainly not alone. There is a huge opportunity in the managed service market, and more channel companies are slowing adding managed services to their existing model. While backup and recovery seems like it would be a staple service for managed service providers to offer their SMB customers, a new study conducted by Intronis MSP Solutions and the 2112 Group found otherwise. Only 29 percent of the channel companies surveyed offer backup, leaving plenty of room for growth in this area.

To help you conduct the most efficient restores and satisfy short recovery time objectives (RTO), we sought advice from Chris Crellin. Chris is the senior director of product management at Intronis MSP Solutions by Barracuda, and he shared some best practices on how you can meet and exceed short RTO requirements.

Best practices for achieving fast and efficient restores

Downtime can be expensive, especially for a small business owner. While downtime can happen for a variety of reasons, it can be costly—about $8,662 per minute on average, according to Everbridge—which creates a significant need for efficient backup and recovery services.

Backup the data your customers need, rather than all of it. The smaller the restore, the faster it will be.  If you’re backing up files and folders and larger sets of data, it might take a while. But, if you only backup what you need, it will be so much quicker.

In my experience, that’s what a lot of our customers do. Instead of backing up everything, they only back up 30 or 50 gigabytes of data the customer needs, versus the entire system. This way searches are quick and restores can be even quicker.

Restore from a local copy when possible. Restoring from a local copy is always the most efficient way to recover data because whenever you need to rely on the internet, it’s always going to take longer. If you’re going to restore from the cloud, make sure you have a good internet connection and good bandwidth pipe. If you decide to forego the extra cost and just have a small 25-megabit pipe, don’t expect a quick restore.

There are a few different options you can use to restore locally, either from an appliance on site or from a local storage system. Making sure you always have a local copy will make your recovery time that much faster. We have multiple examples of folks who only set up cloud backups, hoping to restore 2TB systems within a matter of minutes, but it can take a while for that much data to come down from the cloud. Restoring from a local copy instead can get businesses on their feet faster.

Have a backup plan for your backup plan. I think a good backup plan should have multiple copies of the data. Not only should you have a local copy, but also a cloud copy. If something happens to your customer’s local storage, such as a fire at the business, you have the cloud backup to pull from. Sure, it might be a little slower, but if you’re helping a customer come back from a fire, that is probably the least of their worries. Having more than one extra copy is essential. Follow the 3-2-1 approach with three backups, two local copies, and one in the cloud.

Keep systems and solutions up-to-date. Maintain your customers’ hardware and overall infrastructure. I mentioned having a decent internet pipe, but you also need to maintain their hardware. If a customer has an old server, laptop, hard-drive, memory, etc., it’s going to be slower to restore, and more likely to fail. If they have appliances that haven’t been updated in the past four or five years, the restore is going to take longer because you won’t have the most efficient hardware for the job. Keep in mind, the hardware that you run software on for your customers plays a huge role in how efficiently your backup runs.

Develop, test, and execute your customers’ restore plans on a frequent basis to make sure the backups work. Testing the backup can make or break a successful restore. Sometimes the solution may tell you that the backup is good, but there may be problems on the back end with unforeseen software issues that could prevent a successful restore from happening. Testing the backup set ahead of time sorts out issues before a restore is needed. Lastly, the more you execute the plan, the more comfortable you’ll be with it, and therefore the restore will be faster. You shouldn’t just focus on the efficiency of your software, but also on how efficient you can be from a human perspective as well.

Data can be compromised at any point in time, and as an MSP it’s your job to ensure that your SMBs can get back up and running quickly. While you may need to test restores a few times to make your technicians efficient, following these best practices can help you recover your customers’ data efficiently to satisfy short RTO requirements.

Photo: REDPIXEL.PL /Shutterstock

Lauren Beliveau

Posted by Lauren Beliveau

Lauren is an Editorial Associate at Intronis MSP Solutions by Barracuda. In this position, she creates and develops content that helps managed service providers grow their business. She also regularly writes The MSP’s Bookshelf and our Ask an MSP Expert column.

Leave a reply

Your email address will not be published. Required fields are marked *