Re: Missing the point...
3par with flash is 4kb. See earlier post above for more info.
3par with disk is 16kb
1435 posts • joined 19 Jun 2007
3par with flash is 4kb. See earlier post above for more info.
3par with disk is 16kb
3PAR released something similar about 3 years ago, but it was specific to I/O optimization not data storage, specifically optimizing the I/O to and from SSDs to minimize wear and maximize efficiency, also improved bandwidth efficiency as well.
You could google for "Adaptive Read Caching" and "Adaptive Write Caching" for HP 3PAR for more info on the topic(you could throw in the google search term 'techopsguys' for my blog post on the topic back in 2013).
is pretty old isn't it ? HP has had it with 3PAR for years, and 3PAR had it on their own for years before that. I have little doubt that EMC and HDS and others have it as well.
I don't think it is very common, but the option has been there for a long time.
HP even has a "Facility as a Service" where they will go so far as build you your own data center from the ground up(at the time 150kW was entry level) and operate it for you.
They also have a "Flexible Capacity" program which is sort of capacity licensing for severs/networking/storage(I assume basically anything data center related) as well.
Though I'm sure FaaS and Flexible capacity are not well known, when HP told me about them at Discover 2 years ago I had not heard of them myself.
"built-in, automated data layout and load balancing of all resources[..]"
So what I've had on 3PAR for over 10 years at this point. Good to know I made the right choice in tech.
Does XtremeIO still force scaling in identically sized units of scale? Maybe they fixed that little limitation by this point I don't know.
Because to me tape in general is still useful. The distinction between OFFLINE and OFFSITE. Wouldn't want someone to fat finger a command and delete all data and all backups(or an attacker do it). With so many things driven by an "API" these days it's easier than ever, more so with cloud services.
Having to have a human physically insert a tape into a library in order to access it is appealing to me anyway. Not to say that tape is the only backup, having onsite (online), and offsite (online) backups are still very useful but as a last resort I still consider it a good idea to have offline backups as well, after being written to physically disconnected from any level of automation that would allow it to be destroyed remotely -- whether that is on tape or another medium doesn't really matter though tape seems like the most likely target.
my real user review of the Lenovo P50 is it works ok(about $3600 loaded up including 3 aftermarket Samsung SSDs and 4 year on site support). 98% of the time in Linux(with a windows 7 VM for some work things), 2% of the time in Windows 7(bare metal).
I think if you don't want the crapware you most likely have to pick a small OEM that installs vanilla software. For me it was more important to have on site support so I wanted a big name vendor. Previous laptop was toshiba (bought in 2010), worked fine though support was a year expired and I didn't want to risk my primary work machine being without support anymore.
Vulnerabilities like MITM to me are such a low threat level for most users it's just a knee jerk reaction by people a lot of times. Most technical folks will practice relative safe computing habits(I don't recall being infected with a virus or malware since roughly 1992), and hopefully not using (m)any public wifi access points to reduce attack vectors.
Less technical folks are likely to be compromised 100 different ways before anyone does a MITM attack against them.
I remember my grandfather before he died was apparently addicted to granny porn. He bought a Dell computer, would go to the porn sites, get infected, then after a few weeks or whatever the computer became mostly unusable, he would call up Dell, they would remotely reinstall the system, and he would go at it again(this was about 8 years ago maybe).
Windows 7 OEM end of sale date is coming up pretty quick (end of Oct 2016), and even now it can be difficult to find systems that will ship with it. MS has worked with big players like Intel to ensure lack of compatibility going forward with older Windows systems on new hardware, my company's PC supplier told me some horror stories about their customers rushing to stock pile compatible equipment before the hardware is no longer available(sometimes literally selling out while they had them on the phone).
Is in the others category. El reg's own reporting has violin revenue as a rounding error.
Not likely dell cares much since they plan to have EMC soon.
"We're good for another 12 months"
Lots of security there. Makes me wonder how screwed the other hyper converged vendors are
blame cloud instead?
There were tons of places reporting the FBI found nothing.
Because outside of templates I can only think of one time I have cloned VMs in the past decade(was making 30 VMs to run a burnin test on hardware). (And I have never done a V2V or P2V migration either)
Currently managing around 800 VMs in vsphere ent+
And move to what ? I have never had a facebook or twitter account but I do use linkedin. Though all of my info is "public" on linkedIn(just career stuff) so there is really nothing to compromise data wise(I believe linkedin had me reset my pw back with the original breach(?)).
I don't use linkedin for MUCH(though I am a premium subscriber), it has gotten me tons of career leads over the years(none of which I need right now), and really if just one of those pans out again in the future(little reason to think it wouldn't) it would of paid for itself right then.
In general I'm not a social person so being able to stay "connected" to the people in my career is handy.
I was thinking more along the lines of it shouldn't take much work to block such simple passwords from being used in the first place. I don't advocate requiring really strong passwords for something like linkedin, but people shouldn't be using 1234567 etc (unless I suppose it is a throwaway account or something). Maybe linkedin has already implemented this since this data is pretty old already.
Like my unconverged vmware clusters that run every workload my company runs?
I don't see a workload that these things could not run that my org would ever need.
Converged usually means external storage.
Doesn't seem like impressive numbers to me, is the HCI market that weak? I mean it's not as if this is a new product, EMC has hyperconverged stuff before and no doubt has been telling customers for months this was coming. I'm sure there are a lot of customers that were holding off purchasing knowing this was coming.
(myself I don't care about HCI, I don't even have "converged" systems, more old school unconverged systems)
sure, and you get all of the other pitfalls associated with ZFS along with it.
I think ZFS is a good file system(I even use it for basic file sharing NAS, with real block storage underneath), but I wouldn't ever really consider it as a storage platform. Not even oracle themselves can seem to get it right.
Maybe just to get better pricing on oracle.
I came across this while trying to find out what happened to all the noise salesforce made on postgres a few years ago
My experiences with zfs and corruption were very frustrating. This was on OpenSolaris. I was expecting a simple tool to just clean the corrupt blocks and mount the FS like any other fsck. But no it went to immediate kernel panic and a reboot loop.
With Oracle i would expect it to roll back trx that were bad.
I was at another company a long time ago where we had oracle corruption caused by both controllers in SAN failing at the same time.(the SAN admin blamed himself for misconfiguring the controllers to allow that to happen. To this day I can't imagine why a system would allow you to configure it in such a way) About 20hrs to recover from what I recall. Though we were still seeing the occasional ORA-600 or something error that indicated a corrupt part of the DB more than a year later. They had no really good backups either. They got budget for real standby servers after the incident though.
The tools to deal with zfs corruption were very immature at the time(maybe 4 yrs ago). It seemed to be generally regarded as voodoo to repair zfs corruption. There were things i tried at the time i don't recall what it was a while ago. Nothing helped all data lost.
Fortunately the data was a collection of backups so nothing really was lost other than downtime from the panics and tracing down the cause of the panics.
The system was a nexenta HA cluster that went split brain. Both nodes tried to write to the same volume and bam it imploded. Nexenta support immediately said we were fucked and to restore from backup. They refused to offer advice on the zfs tools used to try to repair the system.
After 2 or 3 incidents of this we disabled HA until we could find a replacement solution.
I'm sure I'm not alone in using linkedin to keep track of business folks and former co workers. Sometimes I may reach out to one after a few years. Not everyone keeps a steady email address.
The other big use case is career opportunities. Though i have to tell recruiters i am not interested constantly I like being in touch with so many people that have expressed interest in hiring me or getting me hired so they can get paid.
About 10 years i worked for a company that was trying to compete with linkedin. I would laugh when i saw my personal network on LinkedIn at the time was more people than my company at the time had in their entire system. Company imploded about 7 years ago.
I almost never participate in the "feed" or whatever it's called the thing where people post things and commment on each other.
I don't use any other social media. I won't touch the linkedin app. Wants too many permissions.
If google didn't do what they did it is likely android would not be in the market leading position it's in.
Apple of course had millions of customers ready to gobble up the first iphones. What did google have? Not much. Microsoft was able to get full control of their phone platform (i think?) But of course nobody uses it.
Webos was the same. BB10 the same. Firefox OS the same (i assume).
For all the doom and gloom about android vulnerabilities I've yet to read about wide scale expoitation. Usually the stories are crap coming from dodgey chinese sites targeting chinese users.
I don't pay close attention most of my tech news comes from el reg. Though I believe if there was serious exploitation going(not involving manually instaling apks or non main stream app stores) on there would be articles here about it.
Every once in a while some security thing comes up that affects a few apps though usually it is yanked from the stores pretty quick.
Android updating won't be fixed in my mind until there is the ability to roll back any upgrade(app or OS) easily. There have been several apps I have updated only to find them worse than before with no way to go back.(I supose one exception is the built in apps. I was able to downgrade samsung S health this way. The newer version didn't support landscape mode and added nothing else that wanted to make me keep it)
System admins etc have a much more secure future than trevor's writing career.
Don't get me wrong we are all fucked eventually (don't care what industry you are in IT or not)
So take the opportunity to have fun while you still can(assuming you can).
Verizon buys yahoo and decides it doesn't want to be in the cloud storage backup biz and axes that offering too. Like they have done with other cloud offerings recently
Can't edit posts on mobile. But wanted to clarify position on ZFS. I think it is a good file system but don't believe it makes a foundation for a good storage platform (e.g. 3par replacement)
I used openfiler a decade ago with some spare HP JBODs for some dev workloads. As long as it worked there was no issue. Forget about upgrading though(I recall the uograde path at the time was basically full data migration to avoid loss).
I tried nexenta a few years ago as JUST a NFS solution for small file set (under 1TB). On paper seemed ok. In reality it sucked hard, made worse by non existent support (yes i paid for their support and professional services to certify the configuration). I have since heard nasty insider stories about ZFS solutions in general which make me glad I have never considered that as a viable solution. I use ZFS at home and it works fine. And I'm sure it has it's use cases at larger scale as well. Keep it away from my vmware and mysql databses though.
Not even solutions like pure storage are mature enough for my liking.
The more i have learned about storage over the past decade that i have been using it more closely, the more conservative I have become in deploying solutions.
I just need it to work without fuss, and for me and my org's mission critical data that means 3par on fibre channel(3par customer for 10 years now ). My experience with 3par is certainly not flawless by any stretch(no solution is perfect). That just reinforces not being interested in taking risks with any other block storage system. My feature usage of 3par is quite limited which probably means I encounter fewer bugs). I love the core of the platform that is very very solid.
Now if HP only had a decent NFS offering (storeEasy and storeAll don't count). 3PAR NFS is a combination of not mature enough and requires special controller versions that only 1 of my 4 arrays happens to have. Even if all my arrays had them I don't believe it would do the job for what I want.
If i was at a larger org we would have more flexibility in testing other things. As-is every piece of storage and server and networking is mission critical(all workloads whether development or qa or testing or production are all consolidated). I have no lower tier of stuff. Maybe at some point but not yet.
For a while we were pumping 200 million in revenue through 8 DL385G7s (384GB with vmware) and a single small 3par array. Today we are bigger for sure. Not big enough to justify segmented servers or storage though.
(one more tidbit, debian user for the past 18 years)
build and install it for you(similarly how some other packages come with scripts that automatically download 3rd party things to install)? just shipping the source sounds like a waste of time, might as well go get it from zfsonlinux.org
But if it is integrated into the system then that would be pretty neat.
I cannot tell if it is integrated based on the article or the link to the package tracker.
(been using Debian with zfsonlinux for a few years now on one system anyway, no real issues)
systems like 3PAR (I think XIV does something similar though last I read they were limited to RAID 1 only, XIO did something similar too they distributed RAID over individual platters from what I recall, so you could have a platter fail and keep using the remaining platters in the disk(systems sold with enough capacity that this was transparent), which is one way they were able to go so long without any disk replacements), distribute RAID across many disks, so you could have dozens or even hundreds of disks rebuilding data in parallel, meaning faster recovery times, and lower latency.
Enterprise class arrays often protect against an entire shelf of disks failing as well(for 3PAR this is the default). So you can lose many drives(as long as they are the "right" drives). Enterprise systems go above and beyond even that and typically only rebuild the data that is written, reducing rebuild times even further depending on how full your system is. I wrote about this more in depth 6 years ago (http://www.techopsguys.com/2010/08/13/do-you-really-need-raid-6/)
Also most/all enterprise systems proactively fail disks before they completely die. Though that certainly doesn't catch all failures, I would wager it accounts for a decent amount though.
SSDs seem to be more optimized for IOPS rather than throughput, short of NVMe SSDs at least, from what I have read anyways SSDs aren't all a million times faster on sequential operations(RAID rebuild) like they are with random operations.
HP at one point said something like 90%+ of the SSDs they have sold were still in operation in the field. Results like this is what helped them decide to offer a 5 year unconditional warranty on all of the SSDs on 3PAR (maybe other platforms too).
my company's first all flash array was a 3PAR 7450 which came online on 2014-11-04, according to the system, from an endurance perspective (initially used the 2TB SSDs, since added 4TB as well), the system is reporting the 2TB SSDs have 98% of endurance left, and the 4TBs installed late last year still have 100% of endurance left. These are for what are otherwise designed to be "read intensive" SSDs, though HP (and others now too) support any workload running on them(HP doesn't market them as read intensive), no restrictions.
I have to askm. W T F are you needing 200 tabs for.
Most i ever have open is MAYBE 25 (across 3 browser windows)
I think much more than that and my Firefox would explode.
I haven't tried on my new lenovo P50 laptop but on my nearly 6 year old Toshiba (i7 dual core with HT). I was forced to restart firefox after maybe 1.5gb of usage. It would just slow to a crawl (64 bit of course). I do have a very old user profile (about 10 years worth of data)
So im sure that is part of it. I don't want to lose the history or cookie access lists (12k+ sites). I know firefox retired the cookie feature i use so not sure what I will do when i get around to upgrading
Even with gigs of free memory. Oh and this is on Linux mint. Maybe it works much better on mac or windows. I have a windows VM but web browsing is generally limited to work related things on it.
I don't use chrome. But if firefox continues to go full retard I may as well switch(to chromium ) because they will be chrome.
To claim that retention bonus
I read a bit of your link and see nothing of interest. What is so interesting to you there. It seems like nothing more than a generic hypervisor running on top of openbsd.
I've been using openbsd as a firewall off and on for about 11 years but don't see a useful role for it outside of that for anything I deal with anyway.
Not that I care about virtuozzo or however it's spelled either.
But it seems folks similar to you are very vocal about every tiny little security flaw.
If security fixes were acutally just fixes instead of new versions changing and breaking other things people would be more inclined to patch more often.
Firefox is like this of course. Windows 10 is like this as well.
Android too. I have been using android 5 more on one of my phones and really dislike it compared to 4.4 on my other phone (phone hardware is identical).
Doesn't say which nation, though one of the quoted tweets mentions london.
I see the linux slack client pretty regularly chew up to 4GB of memory before I kill it. It works pretty well in general. I don't really use any of the fancy stuff(others at my company do) I just treat it as a dumb chat client, since that is all I need.
Seagate has had those for a while both laptop and enterprise
How many windows phone models shipped with intel cpus?
I think the number was quite small if there were any so intel dropping support shouldn't matter MS wasn't using them anyway
is that for the low end model or for the high end model?
Just looking at what HP claims for the 8200:
"The HP 3PAR StoreServ 8200 All-Flash Starter Kit features the same flash-optimized architecture as the entire family of HP 3PAR StoreServ Storage arrays—including cache handling enhancements and system-wide striping to deliver over 75,000 IOPS and sub-millisecond response times. Performance is also scalable to 320,000 IOPS." (I have no doubt the numbers here are quoted at 100% read)
So this 30k number is about 1/10th what the low end 3PAR 8200 can do? (unlike the 8400 the 8200 does not go beyond 2 controllers).
On that same note, assuming the article is accurate I had no idea the existing VNX arrays were *that* slow(apparently not much more than 10k IOPS).
It will be easier to 'root' your phone ? I tried to root my backup Galaxy Note 3 (which runs Android 5) so I could try to downgrade it to 4.4 (really do not like the UI stuff in 5, and when the Galaxy Note 5 came out that prompted me to buy a 2nd Note 3). After a few attempts I gave up, felt like I was more likely to brick the thing then unlock it.
I suppose that is one thing I do miss about WebOS, just had to type in a simple command into the UI and it unlocked it, hook a usb cable to your computer and you get full root shell.
I still have to keep wifi disabled on my main phone almost all the time to prevent AT&T from upgrading it to Android 5.
on prem infrastructure, and SaaS for the rest.
forget IaaS cloud already(and PaaS too)
Looking at the specs for the IBM 900 it doesn't seem on paper to be anything close to DSSD, which on paper is most likely greater than 10-50 times faster. Big part of this of course is PCIe flash, which the IBM Flash 900 does not appear to use(since it only seems rated for 1 million IOPS).
IBM isn't alone here of course, HP has no answer to DSSD at this point either. Though I think the relative market size for a PCIe based flash system is quite small vs the more general purpose SAS flash systems like XtremeIO, 3PAR, IBM FlashSystem, PureStorage etc etc.. Having IOPS in the single digit millions for a single flash array(with 1ms of latency) is going to be enough for just about anyone for a long time to come.
RoR boom to me was in the mid 00s. These days the boom has sort of moved on in my opinion to things like node.js (which is even worse than RoR, hard to imagine they could of gotten worse)
Unless you put it in context as to what market that pay is in.
$105k could practically be considered minimum wage living in the SF bay area anyway (having just left there on Monday).
Also helps to say what someone in IT that doesn't have "devops" might make in the same market.
I couldn't live off $105k today no way.
At least dell will save some cash not having to have so many senior execs.
Lived in the bay area for 5 years(Just moved away yesterday) never once had the slightest interest in attending vmworld. I did attend one vmworld "party" last year(sponsored by 2 companies that I am a customer of), didn't know anybody so just stayed for about 30 minutes then went home. Waste of time.
Conferences have never had any value for me personally at least. HP Discover the past couple of years had it's moments, though outside of meeting with the storage folks whom I know the rest of the event had nothing of interest (not attending Discover this year).
Company more than happy to pay for conferences though none of them interest me.
I have been a loyal vmware customer for nearly 17 years now though. Just vcenter+enterprise plus (5.5U2) though(and vmware workstation). Maybe upgrade to vsphere 6 by end of 2018, no rush.
None of the other products are particularly exciting to me. None of the competition (e.g. openstack, public IaaS cloud etc) is interesting either.
United sent me notes they were putting this in place but I thought it was put in place a month ago. Haven't had to login again yet.
The one that was most scary to me was state farm. Asking me questions like what steet did i live on 30 years ago (i was a young kid, 25 years before I became a customer). The answers were multiple choice. These were records from their databases, they never asked me to setup questions they just asked based on what they knew about me already. Quite startling to me anyway.
I realize insurance companies have a lot of data but did not expect to extend that far back long before I had any accounts under my own name.
I'm guessing they axe at least 20% in the same time period.
I would prefer more advertisers use it, since flash is click to run anyway, I hardly ever see it. Or if browsers would somehow make html5 animations and video etc click to run as well that could work too.
The only thing I can think of that I need flash for on a semi regular basis is Bank of America ShopSafe
it's funny people still think that ala carte services will end up costing less than bundling. As time goes on though more and more are realizing bundling is what saves money (few edge exceptions - e.g. if you are satisfied with the content that Netflix has to offer and just use that for example, though I read increasing comments on people who feel like they have to subscribe to a half dozen or more streaming services to get what they want and that services seem to be fragmenting more as time goes on).
I'm older school I guess. my Tivos have curated (to some degree) what I have been watching on cable TV for the past 16 years now, it's a system I am satisfied with at least. I probably "stream" less than 10 minutes of video a month (generally youtube), haven't streamed music since the late 90s.
I'd say get off my lawn but my apartment doesn't have one.
Would be kind of surprised a serious EMC shop would consider something like this regardless of what it can do.
I know I wouldn't ever put it inline between my systems and HPE 3PAR systems. Though for low end storage like LSI, or maybe HP P2000 type stuff then ok would consider it (but I wouldn't use those storage systems to begin with so once again it's out :) )
(the web client)
.NET client works fine. I hated the .NET client (as a linux user of 20 years) back when I first started using ESX over a decade ago, but this was more of a case of "be careful what you wish for", when the "improved" web client actually seems to be much worse than the existing thick client.
Maybe by the time I do another major upgrade in vsphere (thinking 2018) the web client might be usable.
Till then I use the .NET client on a XenApp (fundamentals) server.