Cloud computing is no longer an emerging technology phenomenon. It’s here to stay and everyone should be considering it when you upgrade or replace your IT equipment. However, many organizations are still hesitant when it comes to cloud computing for fear of data loss or the inability to meet specific performance requirements in a public cloud environment. There are ways to overcome these fears and plan for successful migration into the cloud.
Here are some tips on how you can reduce anxiety about migrating into the cloud:
1. Test before moving completely into production clouds
The only way that you will ever know if a public cloud provider meets your needs is by testing their service with real workloads. Be sure that you collect the necessary metrics to determine performance both during normal operations and while failing over to another cloud. This will give you the information you need to decide if there are any gaps in service that will prevent your apps from running properly in the cloud.
2. Read the SLA very carefully before committing
There is nothing worse than moving into a public cloud environment only to find out that an outage or other problem prevents critical applications from running properly. Before you commit, be sure that the provider has published their Service Level Agreement (SLA) so that you know exactly when things can go wrong and when they must be fixed. Then define your own operational level agreements (OLAs) with their support teams for backup plans when problems occur in production. These contingencies will add cost but in the long run they will prevent major problems when you need those most.
3. Conduct “war games” for business continuity in the cloud
Migrating to a public cloud does not mean that your data is safe forever. Things can still go wrong so you must be prepared for disaster recovery when it strikes. Some disasters are planned events like natural disasters or terrorist attacks while others are unplanned like data center failures, network outages and malicious acts by hackers. You must plan all of these scenarios into your DRP (Disaster Recovery Plan) before launching into production clouds because time is always the enemy during real disasters which can occur without warning at any time of day or night. The best way to test your DRP is to take part in “war games” where you plan for the worst case scenario. No one wants to think about their data center blowing up but it happens more often than most people realize, even in the US.
4. Ensure that your staff is well trained on public clouds
You won’t get much use out of a cloud if your staff doesn’t know how to use it properly either due to lack of training or simple lack of awareness regarding what’s possible with cloud computing. It only takes one bad experience before you lose all trust in any provider so be sure that everyone knows exactly what they are allowed to do and when not to touch anything unless specifically told otherwise by your organization’s IT department.
5. Establish a “point of control” for security compliance
If you are moving to cloud computing, it means that your staff is likely using personal devices which may not meet the same security requirements as your IT environment. That’s why you must establish a “point of control” through your IT department so that everyone knows exactly what their responsibilities are for everything related to securing data, regardless of where it resides.
6. Plan for access to legacy applications in the cloud
Cloud computing has opened up many opportunities for organizations around the world but there are still some instances where you will need to maintain access to legacy apps which might not be available on any other platform than physical servers. This usually happens when an organization cannot find a suitable replacement hosted solution or if they simply refuse to let go of their old applications no matter how inefficient they might be. If this sounds familiar, you will need to open dialogue with your encryption team to make sure that you can modify legacy apps on the fly during failover while still maintaining throughput and security.
7. Know how cloud providers are leveraging new technologies for DRP
Cloud computing may seem like a dream come true but it’s all too real if an outage occurs without warning on any given day at any given time of the year. That’s why every organization should ensure that their cloud provider is using best practices for disaster recovery techniques so that they have the latest technology in place before moving into production mode. You won’t get much value out of having servers sitting idle unless you know exactly when they will be needed most so insist on regular updates regarding their DRP including details about which specific technologies are in place for fail over when needed.
Conclusion:
DRP is essential for public clouds it’s important to remember that no matter how many precautions you take, nothing can ever be 100% safe from data loss during disasters. The best you can hope for is to be prepared with a well thought out Disaster Recovery Plan which should be tested regularly so that your staff knows exactly what to do in the event of an outage.