A qualification methodology
A SearchStorage forum user identified as RK posted this question, which got a ton of responses.
One set seemed particularly apt, and we offer it as our tip for this week. If you have something to contribute along this line, follow this link
Q: I am looking for any info regarding comparison or reviews of Enterprise Storage from EMC, Compaq and IBM. What little I have found so far is not much more than sales brochures content.
A: (provided by another forum reader identified as Devaisha). I didn't find anything myself. Here's [an approach you can take, based on] what I did:
1) Collect all information you can get on storage and systems as now implemented in your company: hardware (server and storage), operating systems, applications, amount of storage used now, amount of storage needed in 12 months, any special problems or wishes concerning storage and backup/restore functionalities, how many and how much of everything is out there.
Try to find out how high your storage costs are, including hardware, software, backup/restore, amount of time spent managing storage (and what is actually done). How much are [you] willing to pay for a Gbyte of centralized, managed storage?
How good are the disaster-recovery processes for all applications, is there anything needed there that could be resolved with new storage features?
Are you experiencing performance problems regarding storage? Are there any problems concerning backup windows? Could you consolidate servers if you had more centralized storage? How reliable are your storage systems at the moment? [Are you having any downtime because of storage (upgrades, migrations, fall outs...)?
How high are your expectations on availability? Would you like to remote boot certain systems, and, if [the answer is] yes, which ones? Do you have central system- management software, and which one is it? Have you centralized backup yet, and, if [the answer is] yes, which product are you using?
Do you really need a SAN or would a NAS be the better first step?
And so forth. The more you know, the more you know what is really important.
2) Classify your requirements (here a few of mine).
Availibilty: The higher, the more redundant and more expensive.
Connectivity: How many different OS do you want to connect to the system?
Interoperability: With other storage vendors, with the backup and system-management products you use.
Features needed / wanted: Remote boot, snapshot, clone, remote mirroring, call out modem, RAID0, 1 to 5.
Performance: Fibre to the disk or just SCSI?
Management: Monitoring, problem recognition, daily use.
Service and support: 7 * 24 ? or less? Technical help and / or project management during implementation?
3) Write a "Request for Proposal (RFP) and send it to all the companies you're interested in. Ask them as many questions as you can and need, but ask them in a that way you can compare the answers (e.g.: do you support remote boot yes / no. Make tables like:
Do you support the following Operating Systems:
NT, AIX, DEC Unix,..;
Do you support following server hardware:
Compaq Proliant NNN, RS6000 H80,...;).
Give them a task to do: How would you solve this situation (and describe something you would like to know); ask them for technical information: white papers an the like; ask them to describe HOW they do something (like remote mirroring, snapshots, clones); ask them about their history in storage and their roadmap for the future; what alliances and partners they have; references. Ask them all you'd like to know! It might be difficult comparing the answers but at least you get hold of some information you need. (And of course, even more questions will arrive: ask them for a technical presentation and get hold of a few server and networking people interested in storage to participate). By this time, you should have a good impression of what to expect and from whom to expect it. By the way, if you use a "standardized" question sheet, force them to use it! Believe me, it's hard enough trying to compare the systems without them sending you tons of shiny sales sheets!
4) Make a "decision Matrix". Take all the questions and answers from the bid and compare them as well as you can. Make a matrix for hardware and software, eg.: max TByte, max. links, max. FC ports, software remote mirror (uses SRFD, PPRC, switched fabric or all components), max LUNs per port, max cache, is cache mirrored? What HBAs do they recommend? Which switch or director? How do they monitor the system? do they support all platforms and applications? ...
Collect all information on management software: what are you allowed to do yourself (interesting when comparing EMC to other companies).
What is absolutely necessary for you (price, availability, connectivity, performance, ease of use)?
What is nice to have?
Editor's note: Devaisha came up with two possibilities for his company. If you follow this procedure, you will likely be able to reach the same sorts of conclusions.
Did you like this tip? Let us know. Email to sound off.
Realizing e-Business With Application Service Providers
Author : Louis Columbus
Publisher : SAMS Publishing
Published : Aug 2000
This book focuses on the market dynamics in business today that are driving the development of the ASP Model and defines the core concepts of e-business. Market needs in all sizes and classes of business are driving the adoption of the ASP model by more businesses than was originally forecasted. Now, the largest software and technology companies of the world are actively pursuing the ASP model as a strong business approach. This book will define the key market drivers behind every aspect of the ASP model, including case studies to illustrate each major component. There will also be case studies of industry leading companies that describe their participation in this market. This book explores Microsoft's Office Online initiative, Oracle's Business Online stores, and the focus of AT&T's and SoftBank's role in this market. This book will delve into the collaborative tools aspect of the ASP model, which is crucial for the long-term success of this marketplace.
This was first published in July 2001