I have just been reading about FalconStor's IPStore product. Now I may be missing something and if I am please tell me, but from what I can gather the IPStor product allows one to transmit IP over Fibre and hence bring together disparate SAN's under one manageable product. Also from what I can see, you are allowing other storage such as NAS to also fall into this umbrella.
Now the way IPStor works is that a Server running IPStor is brought into the network and to achieve continuity of operation they need to be clustered too.
Now bearing the above in mind my questions are:
1. Allowing IP to run over fibre effectively means that you are converting a SAN into a NAS. If I look at Network Appliance, their NAS product does exactly what IPStor is doing with a SAN. Is this too simplistic? Am I missing something?
2. By introducing a further server into the network another layer of complexity is added as well as another point of failure. Surely we need to reduce the level of complexity and ensure that there are as few points of failure as possible.
In other words, why would I want to use a product such as FalconStor's if I can get the same functions with a NAS product?
The FalconStor product is a software based solution that can be introduced into an existing SAN or NAS or both, that virtualizes the storage behind it. The storage can then be pooled and allocated to other servers via block based or CIFS/NFS clients. The block-based clients require a client stub for each server attaching to the SAN and other clients connect via standard IP. Fibre based solutions can run multiple protocols, as it is just a transport mechanism but this software type solution will usually provide access via IP.
What the FalconStor product gives you is consolidated management of both SAN and NAS resources and storage pooling across heterogeneous storage hardware. This approach is being used by many vendors today. Compaq is the first that comes to mind with their Virtual Replicator product. The product enables IP clients to utilize block-based storage access over IP by pooling the back end SAN via software. This enables wide area clients (or local area client) access to disk as if it were attached locally. The benefits to this approach are:
Better storage management
Ease and reduced cost of storage deployment
The caveat here is the need for a fast IP infrastructure for performance and the fact that the storage "server" now can become both a single point of failure and a bottleneck if not configured properly, hence Gigabit Ethernet and Clustering.
For database or any other high throughput clients, plain old SAN is still the way to go. But for those clients who only need file/print or you want to consolidate storage for many servers not requiring extreme throughput (some web servers), then these solutions make a compelling argument.
Dig Deeper on Storage management tools
Related Q&A from Christopher Poelker
SAN expert Chris Poelker compares connecting a SAN with wavelength cabling and dark fiber and discusses the pros and cons of each. Continue Reading
SAN expert Chris Poelker discusses how to change the size of a LUN in a Microsoft cluster server environment. Continue Reading
Storage expert Chris Poelker outlines WWN basics in order to answer the question: "Why do HBAs in a SAN have same base?" Continue Reading