34 posts • joined 14 Aug 2010
Avamar had an issue like this, but more so related to it's Linux OS. Pretty much had to do a reboot each year, until the appropriate patches were installed. A few other EMC products, also running under linux had this issue.
Luckily for EMC, they are aware of which systems and customer are affected, and were proactive about the fix.
This one though, a bit more extreme at only 80 days. Seems like QA as of late is taking a dump here. This with the avamar/datadomain issues, what a pain in the arse.
I didn't see a mention of vSpex or FlexPod? Both allow a customer to use commodity servers, infrastructure, and templates for storage; and build out a reference architecture without the VCE like premium you say costs a few virgins...
It'll still use the big iron of a VNX or FAS filer, but you have a reference template to build off of that is known to work and can save you a bit of money if you don't need converged support.
How is EMC in there as "In use?" It's not even available for sale yet. Unless we're talking an all flash VNX, or flash in the VNX/VMAX?
I think from this reading, it's basically a 3-way tie for the most part, Violin, Pure, and the potential for EMC to make a move.
I know of some EMC demo boxes out in the wild, and I'll leave it at that. I know of a LOT of EMC folks moving over to Pure from the sales and SE side of things.
All EMC needs to do is actually hurry up with a product. Pure is getting the sales team together, from EMC, and needs to get some big wins, and marquee customers. As for Violin, I have yet to see them in my patch of the woods.
Interesting how you would compare a VNX to XtremeIO, when they have nothing in common... Nothing.
I don't see how in the world NetApp thinks ONTAP can do what they want under a single OS. The thought of all these new features being bolted on, patched in, etc; onto an existing platform is a recipe for disaster IMO. It won't perform, won't scale, wont be easy to manage.
There is a reason Isilon is so good at what they do. There is also a reason VMAX/Symmetrix, HDS VSP, HP 3Par, etc are so good at what they do; with purpose built technologies. The thought that a single set of controllers will do all of tasks mentioned under ONTAP, as a single point of failure (like how they try to sell me on using snapshots for backups) It just scares me, and should anyone listening to their fodder.
EMC has a dropbox competitor, it's called Syncplicity. It works with isilon or atmos and soon im told vnx
Going private is needed for Dell
For Dell to survive, they have to go private. It'll allow them to burn through a lot of middle management, shake up the company in terms of products and focus, and generally not be a loss leader anymore.
If sold to Icahn, there will be a payout to investors like himself, higher than what would be received by going private. The company would be put into more debt, and the company continue their downward spiral and will fail. Pieces of the business sold off and massive layoffs. More so then simply a management shake-up from going private.
So on one hand both scenarios get a payout to investors. However the greedy folks always want more. But only one of them will not involve taking on more debt to pay shareholders.
Symantec has been selling Storage Foundation for a decade if not more. They have had a very TIGHT compatibility list of what hardware to use from storage to HBA's to servers. They say "buy commodity hardware, use our software." Well it sucks and can be plagued with a lot of fingerpointing from various vendors. e.g. too slow of disk, wrong HBA driver/firmware, Multipathing issues, etc. This mantra has been used in PureDIsk, NetBackup, etc too. There are other vendors like FreeNas and the like that offer this as well.
What is predicted in this article is years and years off. It's been done for years, but getting it right has never been possible. This is why the HDS, EMC's, and Netapps of the world continue to operate and work. Because their wares have proven themselves. Sure a startup will spin a compelling story, but for what? Some budget conscious admin that likes to be on the bleeding edge? Yeah good luck with that.
How does a benchmark on a system 2+yrs ago, a few models older, a few generations of OneFS older; to a system with an unreleased OS, on a apples to Oranges disk subsystem?
And how relevant or realistic is this benchmark at all? It's bragging rights, but no one builds a system to match this benchmark workload.
I much prefer $.00 per IOP. And not node count or disk config or max IOP's in a synthetic benchmark.
This is great!
I think this is a great idea, that more vendors should have created and marketed. The first pic in the article is so true in so many SMB's.
LANDesk - Great product, poor management
A lot of mid and upper level managers completely ruin the potential that LANDesk has. The "Red-sweater," club as they are known as they are former Novell Employees, and boastful of that. (Not sure why IMO).
That said, it's a great product, that doesn't require much infrastructure, but poor management execution and limited vision and focus has kept them at the desktop level and out of the datacenter where the real money is.
While I've heard "Evil Machine Company," They really are not that way anymore, that's legacy EMC. Where sales reps would strong arm customers for PO's.
Now it's more like "Everything More Complicated," or "Everything More Costly."
You sure about that?
I'm VERY sure there are different sizes of VMAX like the 10k, 20k, etc. I think there is also a specific MSP version as well. The 10k and 20k can come close in price to the VNX 7500 depending on how the discounting is applied. VMAX they can go deeper on the cuts I believe.
Also EMC World will have something new for the VNX... At least that is what the sales reps are saying.
The only folks that liked the product were the Symantec SE's that were fed the kool-aid. Well that and some of the original customers where they still had a fantastic server and desktop backup tool, that got deprecated in favor of Ghost and BackupExec, what a joke and huge step down in functionality.
The product was overly complex, required a lot of server hardware, invasive on the desktop, and very slow to adopt new OS's from MS (It took more than 2yrs to support Win2k8). Then to make matters worse, Symantec was pushing it to all their AV customers and BackupExec installs. Saying that Altiris is the best way to deploy software. NO THANK YOU and good riddance!
EMC has ATMOS for object store. As I've been told by sales reps, object store in isilon isn't going to happen anymore, due to ATMOS.
Re: Good news for SME's
VMware had a backup product before, it was terrible. This new iteration is the replacement. I don't see how VMware is charging anything more than what it did before?
That said, this is also a stripped down version in terms of functionality, so I see this as mainly an SMB play only No replication, no tape out, fixed size, etc... Basically not a fit for SME/Enterprise at all, exactly where the full blown Avamar product plays. Avamar was NEVER an SMB product for obvious reasons...
Find out the market share of GPFS vs StorNext.. StorNext had the mass market appeal and OS support that others didn't. Probably cheaper and easier to use than anything IBM!
Ultimately StorNext is a good product, still in use in a lot of multimedia based shops from broadcasting to game design, movie production, etc.. Slowly though, large scalable NAS systems like Isilon and others are slowly chipping away at their past dominance.
Add to that Quantum's die'ing business in the form of the DXi and tape. They should split StorNext off into it's own business IMO. It's the only one good product they have left.
No it hasn't. Though they were slowly bringing it together in NMM2.2-2.3... As an example, prior to NMM2.4, folks still had to use NMSQL for certain SQL maintenance tasks during the backup.
With NMM2.4, no longer. The code is merged.
For years Overland was in dire need of a cash boost. Their products were never that great, just average. Is this how they'll get their much needed cash? through royalties?
Lots of folks I work with on the West coast in the US use Quest software for SQL and Vmware. From performance monitoring (Toad), the all too common SQL backup tools (Litespeed), and vmware backup tools(vRanger).
This is a HUGE win for Dell. However, Dell's execution to merge products and do it in a timely matter will be the biggest obstacle for them to get over.
Can you imagine how the government uses super computers running racks and racks of thousands of CPU's and TB's of RAM, processing this data in a matter of minutes to seconds? Brings new meaning to "All your base belong to us."
So BOOST and Catalyst are effectively the same in theory. But at what dedupe levels did HP use to calculate 100TB/hr?
Symantec does the same thing with their client-side dedupe, quoting that throughput assuming a 10-15X level of compression.
DataDomain on the other hand is not client direct (well in small cases it is) how about vendors just quote pure RAW ingest numbers?
Tiering is a necessary evil for the next few years.
VMAX/VNX/Clarion all have their place in storage and with their tiering. FAST, FASTVP, VFCache, etc are all cool tech, and necessary too.
EMC does have a "hadoop," like storage and that comes in the form of Isilon and Atmos. Each with respective use cases.
I think you got angry at something EMC did a whiles back and never looked back, all the while EMC has surpassed most big players in the market without you even realizing it with your short-sightedness. Now once the costs of MLC/SLC goes down and reliability up an order of magnitude, we'll have some interesting stuff for storage. But from a general sense, EMC's product line is pretty darn good, if not expensive at times. But no one faults IBM for that, it's just par for the course ;-)
A financial analyst at an investment bank, making deep technical analysis on products he's never managed in his life. Where do I sign up for his newsletter, he sounds uber smart.
I see no way in hell this is an accurate assessment of technology.
All you need is BOOST licensed to get this to work. Of course BOOST can cost a bit on the larger DD boxes. Then you download a plugin on the my.datadomain.com site.
But BOOST is/was already in other pieces of EMC's stack with Networker and integration with Symantec. It can accelerate SQL, Exchange backups directly to DD without touching the media server. Pretty slick.
At the end of the day, looking at the actual data in use at the block and segment level... Does it really matter that you have 500GB, 1TB, or 16TB? No it does not. Only small tidbits of data need caching solutions from a majority view of the market. Sure there is a need in very select few datasets and customers that may need a large high IOP solution, but boasting about 16TB means nothing to the educated storage admin.
It's proper data management, and granular tiering solutions that are needed. The hardware is already here, we need the software to catch up.
Premature don't you think?
Seeing as how the VNX has some of the latest Xeon processor technology, you allude to there being a CPU bottleneck. If you were talking CX, I would believe you, but the VNX less so.
That said, your analysis is flawed in that EMC is typically very tight lipped about products until their release date, so your article is all conjecture and assumption till otherwise noted.
Poor comparisons of competing products
How is the Dell Dedupe box compared to the lowest end EMC DD160? Perhaps from a throughput standpoint, the cheapest DD box can meet or beat it, but from a capacity standpoint, you have to go to a 600 series box. And then, everything is better on the 600 series box in terms of performance, boost or not.
Add to that OST/BOOST is not a Symantec only technology. You better check your sources on that. Look at what EMC is doing with OST/BOOST in their own product lines.
Though I see a lot of clueless SMB's buying these boxes for cost alone and will be happy with their 5:1 dedupe rates. Poor souls.
I've sold a lot of Symantec Endpoint Protection and its NAC capabilities. What made the product standout, was a great administrator that could make sense of the policy capabilities. It's a very good product when configured right. Sadly 80% of installs are done by hacks that are bloody fools thinking they know it all and "it's just AV, how hard can it be?" mentality....
Now I work for a company that standardized on McAfee and ePO, and I noticed a few things I can't do like install CCleaner. It gets wiped within seconds of install. Due to it having a secure erase function. This I found by looking through the logs and registry keys of the app.
The hardest part is that no single product is good enough, and often, multiple products must be used to apply a solid endpoint management and data protection strategy (You see two disparate markets)
You've got desktop management products, AV products, disk encryption, and data management/access/protection products. No single vendor can combine them into a single product, nor would one want to! That is just a mess of complications and agents.
At the end of the day, Security Officers will say you need X/Y/Z feature, and have a list of all required features. But the reality is that, <insert regulatory requirement here> probably only needs 30% that, but it's poor comprehension, or overzealous fools that buy the kitchen sink, and only deploy the part of it.
Side note, complaining of 25Minute boot times? Where are we 1995? Seriously, get modern hardware, software, and updated applications. My guess is that you run 2 versions back on your AV software, still on XP, on a PII machine with 256MB RAM.
Why pick on just RIM?
When iCloud went down momentarily as the new iPhone 4S was being released, no one cried foul. What is the "jesus phone," from Apple immune from criticism? When or if iMessage goes down, I'm sure no one will think twice about it.
As for comments about a "crappy network," that RIM uses for their cloud infrastructure. I'd bet the farm that their core switching costs probably cost more than you'll make in 10 years. Networking is not cheap, let alone one that sees the traffic they do based on their business model. Did you think it was hosted on Linksys gear?
What people are angry about is not the outage. This was just another "nail in the coffin, " so to speak. What people want is more modern phones, more multimedia features, shiny new stuff. Instead, RIM completely missed the boat; focusing on the Enterprise first, and consumer after. Add in outages, a failed tablet launch, very slow OS upgrades, carrier restrictions... You've got a company that has lost focus, consumer demand, and brand loyalty. Remember when Nokia was huge back in the late 90's and early 2000's? They are virtually non-existent these days, or at least not a contender in the pecking order of phone choices.
While I use a BB, It's now a 2+ year old model, it's showing its age. It however serves its purpose as a corporate phone with it's 2nd to none security measures in place. Though, first chance I get I'll probably be moving to an Android model if RIM doesn't do something to entice me with new models and features.
Author is a bit misinformed
Symantec already has a couple of appliances, has had them for a few months. A NetBackup 5000 and a NetBackup 5200. The 5000 has been out for months. The 5000 is a PureDisk Dedupe box, supporting up to 16TB of deduped data, and the 5200 is a Netbackup media server supporting up to 32TB of deduped data. Bot are fully populated, xeon powered boxes with lots of SATA based disk, and RAM in a rack mount box.
Vmware HA is weak
Like the article states, Vmware is only capable of machine failures, a purple screen of death. A BSOD is still an up machine. Symantec has the capability to run SQL queries, or HTTP GET commands, or even check SMTP connections to verify an application is up. This goes much deeper than traditional HA tools from the OS manufacturer.
Not only are there numerous bug fixes to this release, but the new R2 edition of BE2010 also supports things like Sharepoint 2010, and SQL 2008 R2 as well as vSphere 4.1. There are also some neat improvements like guided wizards for setup, better removable drive support, and more NDMP support.
Ultimately, other than bug fixes, and platform support. The changes are more enhancements to simplify setup and administration... As if BE wasn't already simple enough...