Thanks for the first question. You are absolutely correct.
RecoverPoint can now be configured as a software-only solution, a virtual RecoverPoint Appliance (vRPA). The vRPA runs on a VMware virtual machine (VM) for less-demanding environments and is available with VNX. This brings an affordable Tier-1 data protection product to the Midrange saving customers up to 60%.
RecoverPoint now offers a choice to ensure the best value for a customer’s business and application environment. Customers can still choose a physical RecoverPoint appliance (RPA) for maximum performance and multi-site configurations. For small sites or less-demanding environments, customers can choose a software-only virtual RPA (vRPA) option for VNX for a lower total cost of ownership.
Yes, 4.0 allows up to 4 remote copies, or 1 local and 3 remote copies. One of the remote copies can be Sync, and the others (or all) could be Async. Each link is independent so it allows you to access different points in time in different locations - it's really cool.
Interestingly, since our implementation is symmetric, you can also replicate from multiple sites into a single location. This gives you the flexibility to build your multi-site environment any way you want.
Please visit the RecoverPoint booth at EMCworld for more details and to see demos. To specifically learn more about multi-site support, how it is implemented and performance considerations, join the session titled: "RecoverPoint: Optimizing Your Environment for Multi-Site Data Protection".
I understand that VPLEX is a virtualization layer which is capable of consolidating storage devices from different vendors. My question is :
1. Does this virtualization needs 3rd storage vendor support? For example, we want to consolidate backend IBM and HP storage, does VPLEX needs support from them? Or is it transparent to VPLEX?
2. In addition to virtualization, it looks like VPLEX's another bright spot of design is workload mobility. We are now able to move VM anywhere within or even between datacenter around the globe, but without storage movement, the VM mobility is nothing. So what's the main technology behind VPLEX to support such storage mobility along with VM mobility? How to make sure data consistency and at the same retain an acceptable performance?
Thank you so much for your questions. Here are some responses to the questions you have raised:
1. VPLEX has a support matrix which includes EMC and non-EMC storage. You should be able to look at the support matrix to determine if your specific array is supported.
2. This question has multiple parts. So, I will try to separate the pieces out:
a. VPLEX provides its own 'copying' of data to create distributed virtual volumes. This is what allows the storage to be present in multiple locations simultaneously. The presence of this storage allows virtual machines to be migrated from one location to another.
b. VPLEX Local and Metro will respond back to the host only after the data has been written to the array (locally in the case of VPLEX Local and across both sites in the case of VPLEX Metro). This coupled with consistency groups within VPLEX will allow you to maintain data consistency that you seek.
c. The focus of your last question is around performance. There is an excellent white paper (EMC VPLEX: Elements of Performance and Testing Best Practices Defined) that talks about the best practices for VPLEX performance. As a gross summary, performance boils down to correct sizing of your environment for the expected I/O load and ensuring that you have sufficient bandwidth when going across sites (such as in a VPLEX Metro).
Please let us know if there are any additional questions that we could help with.
> What is the advantage of the enhanced integration with VMware SRM for any-point-in-time recovery?
Thanks so much for the question.
With !RecoverPoint version 4, VMware SRM users can designate any of the available points-in-time, to be used once the SRM plan is executed. With this, the full potential of RecoverPoint is unleashed allowing users the same benefits they have as when not using SRM: For example utilizing a "golden" image or an application consistent point-in-time and checking out different points in time for a test/dev scenario.
The point in time selection is done through the EMC VSI extension to the vSphere client (a free plugin).