The Six Differentiating Features of RecoverPointPosted on June 12, 2012 by TheStorageChap in RecoverPoint, Replication
With all of the interest we get in distributed, active-active, datacentres and VPLEX, it is sometime easy to over look the fact that we have one of the most feature rich replication solutions in the market today with RecoverPoint.
With thousands of deployed clusters, millions of run hours, 50 supported platforms and 5-9s system uptime, RecoverPoint continues to advanced through the ranks of replication solutions within EMC, now being positioned for Universal Replication; any application, any storage, any location and any point in time.
So what is that differentiates RecoverPoint from the other replication and recovery solutions available in the market place.
1. Heterogeneous Replication
RecoverPoint is a heterogeneous replication solution able to replicate data across the EMC portfolio of block storage arrays, as well as a large number of other storage vendors storage arrays. This enables customers with mixed storage environments to have a single replication solution to reduce operational complexity or enable tiered replication solutions to reduce overall solution cost. For example a Tier 1 array in Production and a Tier 2 array in DR.
Using the same licensing, customers can set different RPO and RTO policies for different data sets enabling both asynchronous or synchronous replication. In each case RecoverPoint does things slightly differently from the norm. In an aysnc. model RecoverPoint works on the basis of Continuous Remote Replication (CRR) where the RecoverPoint appliances try and drip feed the data across to the remote side as quickly as possible, given the currently available bandwidth and the application write IO; the advantage is minimal lag at the remote site and steady bandwidth utilisation. As well as true synchronous replication, RecoverPoint is also capable of Dynamic Synchronous replication. The problem with synchronous replication is that at times of high load you can impact primary applications because of having to wait for acknowledgements to be returned from the remote site. Dynamic Synchronous replication allows a policy to be set (for example if latency reaches 5ms) so that RecoverPoint can dynamically switch between synchronous and asynchronous replication models negating the effect on host performance.
3. Rollback to any Point in Time
It does not matter if you do synchronous or asynchronous replication, if you have corruption locally that corruption will either immediately or sometime later be replicated to the remote site. Traditionally organisations have used snapshot and clone technology as a method to provide points in time to recover to, but these get taken relatively infrequently, meaning potential data loss in the event of having to roll back to one of these points in time and they can consume larges amounts of capacity. RecoverPoint does away with the need for snaps and clones and replaces them with a Journal system. A bit like a database redo log, the Journal maintains the historical block information for a designated amount of time, enabling the the data to be recovered to very granular points in time, for example every write or every few seconds, vastly reducing any potential data loss. During a 24 hours period there will be thousands of points in time, much more granular than any current snapshot technology on the market today.
4. Application Integration
RecoverPoint has native integration into common applications like Exchange, SQL and Oracle enabling application consistent Points in Time to be created which guarantee application recoverability for that particular point in time. RecoverPoint also supports EMC Replication Manager enabling centralised control of application consistency and application recovery, reducing operational overheads when compared with other solutions in the market place, for example NetApp SnapManager. RecoverPoint is also a Site Replication Adapter for VMware Site Recovery Manager, ensuring support and integration within VMware environments.
5. Disaster Recovery Testing
One of the most difficult aspects of a disaster recovery solution can be DR testing. DR testing can often be a painful process that involves stopping replication, therefore putting the primary business at risk, meaning DR testing has to be done during none working hours and/or having to complete a full resynchronisation when finished. With RecoverPoint you can choose any of the application and crash consistent points in time that are available within the Journal and mount the data at the remote site in a read/write mode, whilst still running in Production and still replicating data from the source site to the target site. Originally designed for DR testing purposes there are a number of other use cases including backup at the remote site and database reporting to name but two.
6. Bandwidth Reduction
Finally, one of the greatest expenses of a DR solution can be the cost of the bandwidth required between the sites to enable replication. Generally the more bandwidth required, the more expensive the link. RecoverPoint includes bandwidth reduction technologies such as compression and de-duplication to vastly reduce the amount of bandwidth required between sites, either enabling more data to be sent over an existing link and delaying upgrade costs, or enabling a smaller capacity link to be purchased in the first place. RecoverPoint also enables many organisations who simply could not get links big enough to replicate using traditional replication technologies to at least have a replication solution to provide some level of protection.