Quick FCoE config overview…

by dave on November 19, 2008

Here’s a quick and dirty overview of a FCoE implementation using the Cisco Nexus 5020 alongside legacy Cisco 9222i’s and Catalyst 6500 series FC and IP switches (respectively).

<content removed for re-factoring; you can still access it below.>
Uploaded on authorSTREAM by dave_graham

Yes, I know that the audio quality is terrible. 😉 I was driving in the car.

Reblog this post [with Zemanta]


Why the AX4-5 matters….

by dave on April 25, 2008

In my previous post, I discussed the reasons why EMC needed to spent a little more quality time with the SMB market at large. Today, in the course of several back-to-back design calls with various level SMB customers, I was struck by a couple of things.

a.) The Clariion AX4-5, while more of a “baby” Clariion amongst its more “mature” CX3 bretheren, offers something that the other arrays currently don’t: Tiered storage within the same drive tray (i.e. SAS and SATA in the same tray). Why is this important, you say? For one, I can now design tiered storage within the same tray, adding value to a customer who previously had to invest in two discrete trays of disk, one for fibre and one for SATA. Simply put, utilize 750GB/1TB drives for the 3+1 vault R5 group, toss in a good 3+1 R5 based on 10k or 15k SAS for a good baseline performance group, and then mix/match other drives to taste. You’ve got 12 slots (wish it was 15) to play around with and, it just works.
b.) You really have a harder time positioning the CX3-10 and NS-2x series because of it. Yes, the NS-22 offers the most versatility, especially with fibre and IP connectivity in the same box, but, again, with all 3 arrays (AX4-5, CX3-10, NS-22) offering 60 drive max with the same performance and storage capabilities, you need to set the table a little more carefully. Obviously, segmenting the AX4-5 into two product variants (iSCSI and fibre) leads to more “purpose built” environment installations as you only can use one or the other. The CX3-10 offers both fibre and iSCSI in the combo model and the NS-22 takes it up another notch by offering FTP, HTTP, CIFS, NFS, iSCSI, and fibre in the same box.
c.) Everyone should be selling the full version of Navisphere Manager with the AX4-5. Honestly, there’s really no reason not to as it does allow for a bit more feature functionality within the box. As a matter of fact, none of my array designs will leave EMC’s dock without it installed. 😉 You’re not missing much with Navi Express, but, once your business grows and you move to commercial Clariion units (CX3-20 and up), you’ll be able to hit the ground running.
d.) Replication, while not as “complete” as the CX3-10 and NS-22 offerings, still covers a majority of the bases required by customers. Mirrorview, Sancopy (important especially with Replication Manager usage w/SQL, Exchange), and Replistor make a compelling entry level replication solution that most multi-site SMB IT shops should be able to afford. Listen, I’m not the money guy so my anticipation of “affordability” might be scoffed at by the market at large, but seriously…finance it. 😉

All in all, the AX4-5 is a powerful offering for SMB and I look forward to bringing more awareness to it in the days ahead.



Seed Newsvine
Add to Technorati Favorites



Seagate sues STEC

by dave on April 14, 2008

EDIT:  4/15/08 @ 827pm EST

After careful consideration, it has become apparent to me that putting myself in jeopardy of commenting on active litigation that (potentially) involves my employer isn’t a smart thing to do. That being said, I’ve decided to leave this content here, but with the following disclaimer:

The opinions expressed here are my personal opinions. Content published here is not read or approved in advance by EMC and does not necessarily reflect the views and opinions of EMC. The information I’ve presented below is through personal research through publically available news sources (news.google.com and other media outlets) and does not represent anything but a high level overview of the potential consequences of this lawsuit. 

Previously, I had written about how Seagate was mulling over the idea of suing SSD manufacturers based on the perceived notion that they had violated their (Seagate’s) closely held patents on how a storage devices is accessed by and passes data to a discrete host. Today, in what I view as a delicious bit of irony, Seagate decided to sue STEC.

In a news entry @ Seagate (found here), they give a little blurb on how STEC invariably violates their patents and how they’re firing the first shot in a battle over IP. The little bit of irony that I see in this basically is this:

a.) Seagate is the largest provider of Fibre, SAS, SATA drives to EMC (and, if I’m not mistaken, other storage OEMs).

b.) Seagate stands to lose a bit of money if the SSDs are rapidly adopted in the EMC Symmetrix platform (and other enterprise class storage arrays, mind you). Now, this may not be appreciable at first, but as SLC/MLC flash continues to drop in price, the pricing battle will become even more noticable.

c.) STEC is the leader in high-speed, high-IOP SSDs on the market. mTron/BitMicro are a distant second, with their primary focus on pro-sumers and early adopters.

What also makes the corners of my mouth tingle a little bit is how Western Digital, with its acquisition of Komag last year, stands to come into a potential power play if Seagate alienates some of its larger OEM partners with this lawsuit shenanigans. Since Komag is responsible for creating the disk platters that reside in most some of the disk drives out there, WD could create a little firestorm of themselves by dinking around with platter pricing.

All in all, this is my take: EMC might be in the unenviable position of paying royalties to Seagate for using STEC’s SSDs in their Symmetrix arrays, further increasing the end-user costs. With this in mind, how about a little “racetrack memory” to solve this problem? 😉

What do you think?