Oracle Releases SPARC T5 Servers; Too Late? 175
First time accepted submitter bobthesungeek76036 writes "On March 26th, Larry Ellison and always with fashionable haircut John Fowler announced the new line of SPARC servers from Oracle. Touted as the fastest microprocessor in the world, they put up some impressive SPEC numbers against much more expensive (and older) IBM hardware. Is the industry still interested in SPARC or is it too late for Larry to regain the server market that Sun Microsystems had many moons ago?"
El Reg has a pretty good overview of the new hardware; the T5 certainly looks interesting for highly threaded work loads (there's some massive SMT going on with 16 threads per core), but with Intel dominating for single-threaded performance and ARM-based servers becoming available squeezing them for massive multi-threading, is there really any hope in Oracle's efforts to stay in the hardware game?
In short, yes. (Score:5, Interesting)
While the T5 may be insignificant to a huge swath of the server market, there are many sectors (financial, energy, Federal, geo, etc.) that make significant use of SPARC platforms. The T5 is a huge advance to these markets. Oracle's not really struggling to stay in the hardware game is the Reg indicates. They produce much of their x86 gear because they use it in the Exa stuff. Their SPARCs are their bread and butter hardware in terms of raw server power. They will sell them as fast as they can produce them. Their recently announced move of manufacturing facilities from Mexico to Oregon is indicative of demand. They build their Exa's in Oregon. They worked a deal with the Oregon state Gov (tax incentives) to move their server manufacturing there in order to compress the logistics lag in getting the servers for the Exa to the kitting facility. Anyway, just my two cents.
Re: (Score:2)
Many large banks are moving from sparc to x64 right now. I know of one major bank that has waited since 5 years for the next generation of sparc, but decided to go to x64 commodity hardware 2 years ago. 50% of servers worldwide have already been migrated.
Probably not. (Score:5, Insightful)
Oracle is going to need to come up with a new game to make waves with the new processor. Simply improving a processor isn't going to change the fact that what people want are low cost processors without vendor lock in.
Sun made a name for itself with interesting hardware, but that was before processing power was a commodity. There are definitely organizations that still run SPARC, and some others who need top of the line performance that will at least give it a shot, but everyone who has a brain and a little industry experience knows that you can't just "try out" the new SPARC with Oracle in charge. If you walk any distance down that road, you start paying premium prices for every little feature you want going forward.
I used to work in exclusively Sun shops, and I've dealt with Oracle for years. There's little that the hardware and their database can do that can't be replicated by x64 and something like Postgres with some thought behind your architecture. For certain, the features they do have are not cost effective against the hundreds of thousands of dollars you pay for Oracle DB licensing, and the premium you pay for SPARC hardware and support.
Re:Probably not. (Score:5, Interesting)
I have not heard one single source say they were leaving SPARC because of performance concerns. The shops that buy SPARC equipment do not have price of servers as a primary concern. Everyone who's left has left because of Oracle.
It is virtually unthinkable that Oracle could or would make the decisions that would reverse this trend.
Sun is on its way out, and SPARC with it. I wonder what Fujitsu will do next?
Re: (Score:3)
I would mod up if I had mod points. This, 1000 times this! Oracle is driving people away due to their business practices, not hardware performance or cost.
Re:Probably not. (Score:5, Interesting)
I have not heard one single source say they were leaving SPARC
Ok , here's one. Albeit a few years ago.. We were having a lot of sad times with the Sun V880. We wanted faster Disk I/O along with a more usable OS. Solaris 9 (& 10), at the time, would boot and run Oracle but it was impossible to get patches for it. We used to download them from Sun's website but then all of a sudden you needed a Vendor ID. After submitting the Vendor ID, downloads still didn't work. iSCSI in particular was important to us but it just didn't work well. Buggy and horribly slow. We finally ended up ditching the V880 and going with two multi-core x86_64 Linux boxes running Centos and SSD raid. The DBA said some of his nightly processes finished in 1/6 of the time it took on the v880. All for a fraction of the cost of the Sun hardware. Yes, the sun stuff is sexy and built like a tank. Yes, it will run for decades without any trouble. If I never see a Sun product again it will be too soon.
Re: (Score:3, Interesting)
Re:Probably not. (Score:5, Informative)
We STILL get that kind of service with our IBM System I (AS400) support.
If you are willing to pay, they have 4-hour support where they will get it there FASTER than overnight. And the tecs are super knowledgable.
Sadly, their blade and x86 support is not REMOTLY as sharp. And with converged hardware it became painful... Fast.
Re:Probably not. (Score:5, Insightful)
wonder why SUN got out of business..
Sun failed because there are two types of people:
1. People who say that good support is really important
2. People that think good support is important enough to pay extra for it
There are plenty of plenty of people in category 1, but few in category 2. Modern hardware is reliable enough and cheap enough that it usually more cost effective to forgo premium support.
Re: (Score:3)
I think you are right. If it absolutely must work at all costs then Sun made sense. Otherwise Cent OS on X86 or something similar is good enough.
Re: (Score:3, Informative)
Assuming a few years ago means 2010 or there about (also "SSD raid" suggests that kind of time frame), so you are comparing a server last sold in 2005 against a five year newer system. Moore's law alone suggests a 6x performance increase. Add in lower latency storage and I'm surprised the processes were only took 1/6th the time.
I admin for an enterprise with a large Sun/Oracle hardware base, and I have serious complaints with Oracle's support procedures, but beyond a few notable missteps, their hardware is
Re: (Score:3, Informative)
A v880 hasn't been current for over 7 years - wtf are you prattling on about comparing ssd raid storage to a box that ran on 10k spindles only?
Load up a pair of T5s with Solaris 11.1, and it will run circles around your gimpy lil Linux setup.
This is doubly true now that OS level optimizations are going in to improve Oracle DB performance...
If you never see a Sun product again, that's because there are no Sun products anymore, they are Oracle products.
Re: (Score:2)
You could at least quote the OP's statement correctly:
I agree, but it looks like an honest oversight to me. He did leave SPARC over performance concerns.
Re: (Score:2)
Where did he say anything about paying? It's just that w/ Sun, he didn't need a vendor ID to download stuff - Sun apparently made it free for anyone who went through the expense of buying a Sun or a V880. Obviously, nobody who didn't buy a Sun would bother downloading that stuff, but that, for whatever reason, didn't make sense to Oracle, since they started requiring Vendor ID numbers. And as the GP pointed out, they ultimately did get and provide the Vendor ID, that didn't fix the problem either, so to
Re: (Score:2)
Re: (Score:3)
We're leaving Sparc machines because of cost. We have to run Oracle because we're using Oracle db/middleware on our ERP, but we've found that Dell's run circles around anything Oracle has sold us in the past for far less money.
Re: (Score:2)
We left because of Oracle buying Sun, but when we moved to Ubuntu VMs we discovered they were ridiculously faster (turns out when you're running a pile of Java, it's actually all about the bogoMIPS), and if we'd realised just how much faster we'd have moved earlier.
Re:Probably not. (Score:4, Informative)
I'd argue that the entire move to Linux is mainly people leaving SPARC because of performance concerns. In the mid 1990s through early 2000s Linux/x86 was a low end system and Solaris/SPARC was the big brother. The lack of performance is what made Linux thrive. If Sun workstations were still say 20-100x faster than x86 workstations and in the $10k range I'd be they would still be selling. If a $40-200k Sun server would crush a rack of x86 boxes they would still be selling.
Re: (Score:2)
As usual, Intel's manufacturing processes won the day in the end. After all, once Intel switched to the multi-core CPU architecture, there was little that they couldn't do in terms of performance by simply tossing more cores at the problem. Who exactly manufactures SPARCs for Oracle today? The ones that did - like Ross & Cypress - Ross died a while ago, and I doubt that Cypress is still in the SPARC manufacturing business, or is it?
Otherwise, if Sun/Oracle had a competitive manufacturing process, o
Re: (Score:2)
Sun lost the battle well before Oracle IMHO. 64 bit sparcs came out in 1994. You could do things on a Sun you simply couldn't do on a PC. Sparc 3 is a cool chip Intel doesn't sell a 16 core 128 thread CPU even today. I'd assume they won't till about 2020 at the earliest so a 10 year lag. The problem is the chip is so expensive that you are better off just buying a more expensive motherboard and multiple CPUs.
I guess the big question is whether ARM is going to do to x86 what x86 did to MIPS, SPARC, Powe
Re: (Score:2)
We left Sparc/Solaris eons ago because of performance issues. It may be true that nobody leaves Sun/Oracle in 201x for performance reasons, but only because those who cared about performance left a long time ago (us in 2002-2004 timeframe)
In fact, they've probably left in stages. At one point, Sparc was one of the "easy" paths to 64 bit. 64 bit Intel/AMD chips ate up that market a long time ago. Virtualization helped with some of the hardware partition issues. And you can get super hardened multiple power s
Fujitsu's CPU options (Score:2)
Fujitsu owns the SPARC64 architecture, so in the end, if Oracle drops it or diminishes, Fujitsu would be left as the default owner of the SPARC architecture. They have too much invested in SPARC to abandon it. Although they could go for Itanium, POWER or MIPS if they are for any reason forced to abandon SPARC.
But I don't see why that will be necessary. With current process nodes and technology, Fujitsu can simply take the latest and greatest CPU they have, make a multi core design out of it, and use it
Re: (Score:2)
It is not trivial in terms of cost for a company to suddenly switch all their servers to a totally new platform. While your decision was your prerogative, you did have choices that would have enabled you to use those SPARCs w/o being stuck to Oracle (except for hardware). You could have migrated existing equipment from Solaris to FreeBSD, thereby eliminating the Oracle software lock-in, while getting new Xeon boxes, putting FBSD on them and running them along w/ it.
Of course, if the client was running O
Re:Probably not. (Score:5, Insightful)
Oracle is going to need to come up with a new game to make waves with the new processor. Simply improving a processor isn't going to change the fact that what people want are low cost processors without vendor lock in.
"People" are not Oracle's target market. Their target are are huge 24x7x365 enterprise clusters that are not about to change databases unless they absolutely have to so for most of them the question isn't whether they'll run Oracle it's what they'll run Oracle on. Whether it's SPARC, POWER or Intel's E7 Xeons with RAS features they'll be paying blood for the hardware, ARM and Postgres isn't even on the radar. If you run a tiny, non-performance or non-uptime critical Oracle DB it's because you're an Oracle shop and have standardized on it, not because you need it or you have a PHB who insists on Oracle because it's enterprisy. I'm feeling pretty they'll sell if Oracle just spins it right that they take full top-to-bottom responsibility (just not liability) for the stack working at optimal performance.
Re:Probably not. (Score:4, Interesting)
And I was very clear that there will be organizations that will continue to run both Oracle and Solaris and SPARC. I have direct experience working at those financial and governmental organization and I do fully understand that they have both a great deal of inertia, as well as a decision making process that is filled with bureaucrats and PHBs.
Still, don't be so sure that the field won't change even there. I don't think I have ever worked in a shop that wasn't 24/7/365 and I have worked in the huge enterprises you are talking about. Sure, they may be running SPARC and Oracle for the next 20-30 years, if they can, just like they ran mainframes almost that long for certain tasks. New growth and new money, however, does not have to take that path, even in big shops.
And this will sell, but will it "save" the line? I don't think so. If they are talking about continued existence of the line as legacy into the distant future, the lock-in has already achieved that more than the T5 or any upgraded processor ever will. Now, if you are talking about "saving" the line in the sense of it returning to a vibrant growth platform beyond its big business/government niche, I'd say that it is too little, too late, on it's own.
Re:Probably not. (Score:5, Interesting)
Oracle is going to need to come up with a new game to make waves with the new processor. Simply improving a processor isn't going to change the fact that what people want are low cost processors without vendor lock in.
They are relying on Oracle Db dominance to bring in the T5. They are working on adding "software on silicon" to future processors so certain DB calls happen at HW speeds.
As long as Oracle DB has market dominance, then anyone who needs to squeeze absolute performance from their DB; then they logical choice will be SPARC.
With that user base they can move in to other segments.
I have nothing against "software on silicon". However, it does smell of anti-competition... I doubt Oracle works with other CHIP designers on this HW API... but I could be wrong.
Re: (Score:3)
IBM was allegedly working on the equivalent of "software on silicon" as far back as the late 1970s ("Future Systems"). Didn't happen, although some of the extremely CISC-y instructions in the current zSeries set aren't too far removed.
One problem with doing database in hardware is that it's a lot faster and easier to modify software than it is to modify hardware. Especially once it's commodity stuff out in the field.
How much truly high-performance stuff is done on Oracle DB is unclear to me. A lot of the bi
Re: (Score:3)
The question is: will doing the DB in hardware even help with anything? I presume it's not that hard to saturate the memory bus on any modern server while doing a database query. If database bottlenecks were actually the CPU power, not memory bandwidth, then it'd make sense.
Re: (Score:2)
Oracle is quick to mention the benefit of hardware encryption they've implemented that's transparent to the applications accessing the data. This is pretty huge because just about every SaaS I've ever worked on has been cobbled together without encryption in the original design and then later on when it was determined to be a valuable feature, the challenge of implementing it in the existing code base was gargantuan. Transparent encry
Re: (Score:2)
I have nothing against "software on silicon". However, it does smell of anti-competition... I doubt Oracle works with other CHIP designers on this HW API... but I could be wrong.
I would ask the question the other way, would other chip designers be interested in "do Oracle stuff to database" instructions on their general purpose CPUs? I doubt that, since dead silicon is would greatly cut into the margins of all their other sales.
Ellison: "Sparc runs Oracle faster than anything" (Score:2)
Not many people noticed this [theregister.co.uk]:
Re:Probably not. (Score:4, Insightful)
Oracle is going to need to come up with a new game to make waves with the new processor. Simply improving a processor isn't going to change the fact that what people want are low cost processors without vendor lock in.
What individual users want is low cost without vendor lock in. What ENTERPRISE wants, and the market for Oracle, is a rock solid platform with excellent support and maintenance. Sun provided that at a reasonable cost. Oracle is simply charging too much for the same product. For example, they've completely overhauled their support costs to ream their existing Sun customers, and they (read we) are looking for other solutions. The company I work for has probably bought its last Sun/Oracle server.
Re: (Score:2)
I recall when EBay first moved to Sun/Oracle several years ago, shortly thereafter they were down for a day or two. As I recall the loss of income was several $million. That's the kind of number that makes the CIO not sleep at night, and lose hair.
Re: (Score:2, Interesting)
Re: (Score:3)
You forget that a T5 has not only one processor but dozens.
You forget that the Oracle DB and Solaris and the SPARC Processor are synergizing each other.
Oracle, the DB, outperforms on Sparc architectures everything else, haedware wise and money wise. The reason is the superb multithreading/processing possible on SPARC hardware.
You want to do that with x86 and PostgreSQL?
Sorry, you must be kidding.
Ah, you also forgot: SPARC is an open architecture. There os no vendor lock in. And AFAIK Open Solaris is not dea
SPARC options (Score:2)
The SPARC consortium has its members, but other than Fujitsu, who makes SPARC computers any more? I do think that any company that does ought to go w/ FBSD, OBSD or Linux, so that their fortunes are not tied to anything Oracle does.
Open Solaris, in terms of this discussion about the SPARC, is indeed dead. OpenIndiana does not have any SPARC support, while a new fork from Open Solaris - Schillix - which does plan to support SPARCs, has just gotten started. So better not go the OpenSolaris/OpenIndiana/Sc
Re: (Score:2)
I'm not sure what SPARC hardware does today comparatively. But Oracle database vs. Postgres. Yes there are huge differences.
Materialized views are a must for views + performance.
Flashback queries and flashback archives are a must for any kind of rollback.
The ability to divide queries across CPUs is vital for complex data warehouse
etc...
They aren't really in the same league.
Re: (Score:2)
Simply improving a processor isn't going to change the fact that what people want are low cost processors without vendor lock in.
This is what they say. Then they put Apple on desks and in bags, and buy iPhones, not because it's lower cost but because it's viewed like a perk (like free lunches, snacks, or an on-site gym, etc) - and it's relatively cheap compared to payroll costs in general. What they say is largely irrelevant because they don't know what they want, and in the absence of wanting something else a purchaser will say they need lower cost. When a purchaser says they're cost sensitive it really says they don't know what
Re: (Score:3)
Thanks Larry. How's the yacht?
it's a marketing problem (Score:4, Interesting)
The performance is just a factor, they can sell if Oracle prices it right, accounting for performance-per-watt of their stack vs. the competing ones.
Sparc being an exotic arch cuts both way, you sure have more trouble with ports, OTOH hackers have to adapt their tools to penetrate those servers, in many cases it's overall a plus.
The main obstacle IMHO is that those servers come from "we are indeed evil" Oracle ;)
Re: (Score:2, Funny)
they can sell if Oracle prices it right,
The right price for Oracle translates in whatever it takes for Larry to buy another island [forbes.com]
Re:it's a marketing problem (Score:4, Insightful)
"We are indeed evil" is the best phrase I've ever heard to describe Oracle. Their mafia-style "well, how much you got?" pricing schemes are insane. Much like the mafia they pretty much force you to give up financial statements and sensitive proprietary business info, then they charge you a percentage of your gross based on how much they think you can afford and how much they think you depend on their software. (Granted they don't admit to this, but this is the net effect of their practices)
I shudder to think what strings they put on hardware they sell to you.
Re: (Score:2)
They will certainly be trying to move from selling to long term leases in the future.
Re: (Score:2, Funny)
The main obstacle IMHO is that those servers come from "we are indeed evil" Oracle ;)
Google: Don't be (seen as) evil.
Oracle: Fuck PR, we're evil.
I honestly don't know which motto is better.
Re: (Score:2)
At least Oracle is being honest with everyone about being asshats.
Re: (Score:2)
Re: (Score:3)
Re:it's a marketing problem (Score:4, Informative)
Their "small" T5-2 has 32 cores running at 3.6 GHz with 256 threads and 256 GB of RAM. A similar and cheaper x86 system doesn't even exist.
Re: (Score:2)
So, just as an example, the IBM x3750 M4 doesn't exist? It has 32 cores and can go up to 1.5TB of RAM. It is less than $50K.
Re: (Score:3)
According to IBM's website [ibm.com], $45k will get you:
4x E5-4650 @ 2.7GHz (32 total cores)
192GB RAM
0 HDD
2 1 Gbps NICs
At the $53k price point the SPARC box also includes a pair of boot drives and 4 integrated 10 Gbps NICs and an extra 64 GB of RAM. More PCIe 3.0 slots. Not to mention the 16x SMT that was already pointed out in another reply. I will note that I did erroneously double the number of cores in the SPARC system. There are only 16 cores.
It doesn't seem to be that the pricing is all that bad comparatively.
Re: (Score:2)
What nonsense. Every hear of IBM's Remote Supervisor Adapter?
http://www-947.ibm.com/support/entry/portal/docdisplay?lndocid=MIGR-50116 [ibm.com]
It gives you a virtual serial console and much more. Simple serial port access is handy but the x86 vendors aren't so stupid that they make you attach physical keyboards or even KVM concentrators. I've worked in environments where dozens of xSeries servers are provisioned daily without a single keyboard or monitor anywhere in the datacenter.
Re: (Score:2)
RSA was just an example. Each vendor has their own version.
DHCP with addresses pre-assigned based on MAC address are used on the management network. Each of the multiple DCs is staffed only with hardware monkeys. Because of security requirements no one with administrative access can even enter a DC. So once the hardware is racked and powered on it could be accessed and configured. We don't do it that way though. We have orchestration software that does all the configuration dynamically. It's completely push
Re: (Score:2)
When I make server purchase decisions, I don't care to talk to anyone at the vendor. I want to read their collateral and make the business decision. If their documentation or collateral is lacking, I won't be prying it from their dead hands -- it's their fucking loss, not mine. Resellers are, I think, a waste of resources. If the list price is there to discourage purchase, then what the heck, I'll just follow their guideline and not buy. Simple as that. I guess it's time to short sell some Oracle stock.
Re: (Score:2)
Re: (Score:3)
It's their problem. I tell all my vendors: if the prices available online are not competitive, I'm not looking any further. Makes it easy for me and them. For me, because I don't have to pester some clueless sales rep about pricing -- it's a waste of my time. For them, because if they want to win my business, they don't have to do anything extraordinary: just provide competitive prices, and I'll make sure the best deal wins. Easy as that.
New Intel 8088s while we are at it? (Score:2, Interesting)
I have not seen Sparcs in years. They are so 2003. Kind of sad as we move to generic x86 but Sun really screwed up marketing these and Oracle is not helping by requiring an Oracle RDBMS license whether they are an oracle shop or not does not help. Oracle has also been happy to tell people with perfectly good Sparc Ultra I's to go fuck themselves we wont patch your systems anymore unless you pay us $$$ for your 12 year old systems you already paid for!
You can tell I do not like Oracle so consider my opinion
Re: (Score:2)
No, SPARCs have been coming out on a regular basis. My biggest issue was Oracle dropping support for so many recent processors with Solaris 11 (UltraSparc IV and older removed, no more IA-32 either). A Sun Enterprise M3000, which was being sold in late 2010, doesn't run Solaris 11.
I do however still enjoy playing with new SPARC systems when my client buys one.
Re: (Score:2)
Um, yes it does. [oracle.com]
Re: (Score:2)
Interesting, I might have to contact Oracle support on that one (when I tried to install Solaris 11 on a client's server, I got a "your system isn't supported" error.
Re: (Score:3)
Interesting, I might have to contact Oracle support on that one (when I tried to install Solaris 11 on a client's server, I got a "your system isn't supported" error.
Support contracts have to be paid, or that's the answer you get.
Re: (Score:2)
Not what I meant - the contract is up to date. I booted to the Solaris 11 disk for sparc (Stop-A, boot cdrom), and the OS threw an unsupported proc error. Client decided it wasn't worth looking into - since everything else they have is on Solaris 10, and decided to stay on S10 rather then spend the time on contacting support, etc.
Re: (Score:2)
I ran into a similar problem years back, trying to put Oracle on a RHEL box....I found the file that the installer was looking for to get versioning info, and changed that to something it liked, and install went perfectly.
Might be something that simple on your problem?
Re: (Score:2)
No, SPARCs have been coming out on a regular basis. My biggest issue was Oracle dropping support for so many recent processors with Solaris 11 (UltraSparc IV and older removed, no more IA-32 either). A Sun Enterprise M3000, which was being sold in late 2010, doesn't run Solaris 11.
I do however still enjoy playing with new SPARC systems when my client buys one.
Can't those older systems - or even the new ones - have either one of the BSDs or Linux (Debian, if not OEL) running on them? Once one has that going, one is totally free from any blackmail by Oracle.
Re: (Score:2)
Yes they can. My client that's using them, however, wants Solaris on them because their client is using Solaris (Oracle DB on Solaris 10). The Solaris 11 idea was to have them be used to using it (it was an extra system) to be familiar and ready to support it when their client moves to it.
Also a note, in college we took some SPARC servers in college and installed Red Hat on them. Ran much better with RH then Solaris (I think they had S9 on them before).
Now that I'm thinking about it, I think it was a dif
Ship has sailed (Score:2, Interesting)
Unfortunately, even in (sizable) niches like telecom, the days of exclusively SPARC shops are long over.
There will be some markets that continue to use Oracle hardware for business continuity sake (Sun/Oracle has ridiculously long hardware lifecycles by industry standards). But as a mass (server) market influence, I think Oracle is done.
Re: (Score:2, Troll)
Unfortunately, even in (sizable) niches like telecom, the days of exclusively SPARC shops are long over.
And there's the rub. What's Solaris got that nobody else has got? dtrace, zones, supported zfs. All of those things are coming to other operating systems. Even if they weren't, nobody is sitting at a Sun machine any more (statistically) so the homogeneity argument has also been lost. If Sun had ever taken Solaris-x86 seriously then that might not have wound up being the case, but they didn't, so it did.
Re: (Score:3)
I think the larger issue with Solaris is that Oracle is intentionally murdering the "mindshare" of their users.
At least part of vendor support is aftermarket support .. right now you can't get patch clusters, bug reports or documentation without a current support contract.
Same with downloading of Solaris media .. if you want to run solaris sparc on your old blade 1500 to compile/debug stuff before moving it to a production machine .. you can't, not without a hardware contract .. at least not easily.
Oracle/S
Java (Score:2)
Re: (Score:2)
That's a good point. Java is definitely in demand and wide use.
Missing the Point (Score:2, Insightful)
The posts so far are missing the point. The point is that Oracle certifies their products to run on their hardware. They have a captive audience.
We're running away from SPARC as fast... (Score:5, Informative)
We're running away from SPARC as fast as we can.
Our unix shop used to be primarily SPARC-based, but with limited IT budgets, we're able to do far more with much less money using HP blades running CentOS.
For most purposes, SPARC hardware is far too expensive and Oracle seems to be doing all they can to kill Solaris.
We still run a handfull of SPARC systems that run specialized applications and a few Solaris zones, but nearly all other services have been pushed to natively hosted Linux systems, or virtual machines running Windows or Linux.
How much is Oracle Solaris? How much for Linux? (Score:2)
Oracle charges a big fat $0 dollars for their Linux port, including free updates.
Oracle was kind enough to drop free updates from Solaris, drop Ultrasparc support, drop OpenSolaris, and by extension drop the userbase. Why pay when a sibling product is free?
Oracle intended to drive customers out the door. What other conclusion can there be?
Re: (Score:2)
Re: (Score:2)
A majority of our IT HW budget is for High Performance Computing. We have about 10000 x86 cores running CentOS, about 100 M2070 GPUs and close to a petabyte of Isilon cluster storage in production right now and will be expanding to over 15000 cores in the next year.
If we wanted to use SPARC systems, we couldn't afford anything nearly as powerful or as painless to manage. We don't need OS support other than patches and we're not tied to any particular vendor (other than Isilon). We may implement a Gluster st
Re: (Score:2)
One is a ridiculous number? Wow. You have very high expectations for ease of use.
I'm guessing that in your IT world the only value proposition is how much horsepower can Intel deliver because everything else is going to be the generic setting. And to be honest it's a pretty worthwhile stance to take, but there is still value in having som
OpenSource Fud.Oracle is not dead to the unbiased (Score:5, Insightful)
A saying was told me to growing up, "Use the Proper Tool for the Job" which varies person to person, BUT for me SPARC and Solaris is the right tool. I see the OpenSource community as a great community. My WHOLE stack runs on OpenSource software. I beta-test/develop MUCH of my stuff on either Linux or OSX.
But when it comes to the production OS, I'm not some blanketed Linux bigot. I'm an *NIX Admin and an Architect at heart.
Professionally I'm a CTO (I do everything from programming php / data-center / network / DBA / UNIX / security / etc.) for an internet-based start-up that runs Solaris 10 and used SPARC CoolThread hardware in production. Baffled why? For a few reasons:
When I did a cost analysis of my time & the company's money vs Intel offerings and SPARC I eventually came away with these main points.
1.) SPARC hardware is still WAY superior with remote management than any x86 POS I've ever managed. The ALOM on a SPARC and a serial cable from my Mac works EVERY time. When I worked in past shops managing thousands of Linux Dells and HP's we had nothing but issues with ILOs from the hardware and OS side. Just pure donkey shit.
When you're a start-up buying used hardware it is a great way to cut cost where investors/owners LOVE. Frankly SPARC hardware in my experience can keep on chugging where those HPs and Dells are falling apart right and left. I don't have time to be fucking with hardware when I'm running the show of a million hats.
2.) LONG-term stability with Solaris 10 and maybe Solaris 11 (still evaluating) is a necessity to me. I work for a crazy ass mad-scientist type who does EVERYTHING custom. He's worse than the scientists that I worked with back at JPL-NASA. He has software that's been running for a decade, and the software/application I write with him now he wants to work years down the road as well. That means, I don't need to worry about a yum or apt get update that blows away some part that is critical to ONLY us and I gotta figure WTF happened. The OS is a critical back-bone element where I've seen "Linux dependency hell" fuck me so many times and cost me so many hours, that I PREFER building my own Solaris 10 packages and Solaris 11 (still in testing for me) packages (Yes, I'm a REAL UNIX admin no these lazy wanders) without worries that the OS will be compromised by something lame. In the long-run I have more freedom to enjoy time with my doggies.
When you work for a company that builds custom crap that. Everything it talks to regarding the OS needs to work without question. I have always have had that with Solaris SPARC and with Support till 2018 or extended 2021 by then I should be retired from the gig! But I KNOW nothing funky will happen with the OS while I'm working here. For each new x86 hardware update for Linux, it's a whole new 'testing' to make sure it doesn't blow up the OS on the next reboot. Never had that with SPARC of maintained properly.
With that long-term support and marriage to the hardware I know the relationship is TIGHT, that can be VERY useful when you're concerned with down the road support or integration. Dell or HP does a hardware update and the RedHat or Debian kernel or images haven't been added, then you gotta do a post image. FUCK THAT NIGHTMARE! SPARC WORKS end of story.
3.) Threads! NOTHING compares to SPARC when it comes to multiple threads and what not. My T2000's running 32 cores make damn good web-servers. They also save space in the rack as well!
4.) Virtualization is WAY superior than KVM or VMWare. I've used many of the OpenSource VM solutions and frankly non compare to the control that I can do with either LDOMs or Solaris Containers/zones.
5.) ZFS yeah, Linux we hear your promises of a bad-ass filesystem, I'm still waiting.
So, is Oracle and SPARC dead? Popularity may go down, that's normal, but it's not "dead" to anyone who has a reason/purpose to use the OS/hardware offered.
The world isn't one big LAMP stack.
Again, I'm not *against* Linux, I use it for development and personal shit all day. However, I'm not a blind follower either.
Does it really matter? (Score:2)
To me database is just a resource, so I may be looking at this wrong, but it seems that the Oracle server is just an appliance, and it doesn't matter to me what architecture it runs on as long as it performs well. So buying an Oracle server on Sparc isn't really a matter of "it's not x86" or "it's not Power". Since I'm not going to use the box for anything except Oracle, the cpu architecture is immaterial. Not even the OS is important, as long as it works well.
Re: (Score:2)
That underlying architecture affects how likely it will "perform well." There's a reason people tend to deploy server applications on UNIX systems rather than Windows.
Re: (Score:2)
That underlying architecture affects how likely it will "perform well." There's a reason people tend to deploy server applications on UNIX systems rather than Windows.
Agreed, which is why our back end servers tend to be Unix or Linux. But in this case, Oracle owns both the architecture and the underlying OS, so one would think that there was some optimization going on.
Re: (Score:2)
Should a server really be just an appliance? Servers must be kept up-to-date with the latest security updates, etc. Appliances are plug in and forget.
In addition, at my day job, we often re-purpose servers for various uses. A non x86 server would severely limit our options when it comes to this. For example, when we outgrow our current database server, we can purchase a new one and utilize the old one as a standby web server, development server, or something similar.
Re: (Score:2)
Should a server really be just an appliance? Servers must be kept up-to-date with the latest security updates, etc. Appliances are plug in and forget.
In addition, at my day job, we often re-purpose servers for various uses. A non x86 server would severely limit our options when it comes to this. For example, when we outgrow our current database server, we can purchase a new one and utilize the old one as a standby web server, development server, or something similar.
An Oracle server, or a Netezza server, or a Google appliance, or a storage appliance, are purchased for a particular purpose. They're not general purpose servers, and as such, it doesn't matter what they run (assuming the admin tools are adequate to whatever administration is required) and it doesn't matter what architecture they use, as long as it's adequate to the task. That most appliances of this kind run some variation of Unix is a plus for old time Unix admins like you and me, but it's not really pa
8MB L3 cache? (Score:2)
To recap, the T5 chip has 16KB of L1 data cache, 16KB of L1 instruction cache, and 128KB of L2 cache for each of its sixteen cores, plus an 8MB L3 cache that all of the cores share.
8MB L3 cache for 16 cores?!
Are they kidding??
IMO: DOA.
Still some valuable niches (Score:2)
Funny thing is, just a couple days ago I was reading a presentation about the obscene amounts of money MorganStanley was saving by switching their OpenAFS cluster from Linux over to Solaris on SPARC.
The advantages are mostly around ZFS, plus dedupe and compression, but they throw in DTrace as well, as having helped track down performance issues with OpenAFS.
http://www.ukoug.org/what-we-offer/library/openafs-on-solaris-11-x86-robert-milkowski/ [ukoug.org]
This is decidedly one place where Linux is lagging massively behin
Oracle wants too much money... (Score:2)
Re: (Score:3, Insightful)
Re:Who cares? (Score:5, Informative)
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
You forgot to say it's in the cloud.
Re: (Score:2)
With the advent of cluster file systems, you don't have to pay for unreasonably expensive "bulletproof" hardware anymore. You can set up a Gluster storage cluster on commodity-grade X86 hardware get all the speed and redundancy you need (and are willing to pay for) at vastly cheaper prices. For those that don't want to roll their own, you can use commercial storage clusters by Isilon or storage virtualization devices such as the F5 Acopia with pretty much any storage underneath that you like.
Re: (Score:2)
clustered file systems are decades old, this Gluster of which you speak is not suitable for serious financial computing as no mature software exists for it. and the company that makes it only has 60 employees, they are going to support a fortune 500's accounting wares? nope...
Re: (Score:2)
Clustered filesystems were always very brittle in the past:
- a few controllers ran everything (single point of failure)
- limited access to them (special APIs you're expecting)
- difficult administration adding nodes and properly replicating
Advances have fixed these in Gluster, but it's new. There will be no "targeted software" since it's a filesystem. And as open-source, a 60-person company can easily produce something consumable by a huge company since they can share support effort.
Re: (Score:2)
not at all, clustered filesystems talking normal system create/write/read have been around for decades. For example AFS, with as many "controllers" as you want, no special API, and easy admin and replication. sure, write performance slow, but it's a robust filesystem 30+ years old.
Support to a large corporation for mission critical apps means much more than just having some forums to answer questions, there is no way a small company could hope to provide such.
Re: (Score:2)
I've worked with sparc machines for decades, Sun/Oracle is losing market share in SPARC realm, less than 5% now. it's dying, write it off, very little place in server world any more. Virtualization and IBM's various PPC architectures and x86-64 is eating its lunch. When an x86 server has a failure no one gives a shit, we pop the blade out, fix it, and slap it back in to join the virtualization cluster.
Re:High priced, no support and little imagination (Score:5, Insightful)
You seem somewhat ignorant of the platform you are criticizing. Yes, we know it's not x86 so the fact that it doesn't run Windows isn't a surprise. You can run Linux on SPARC, don't know why you would want to though, Solaris is very good. No, it's not Linux, but it's still very good.
As far as virtualization goes, they've had hardware support for longer than the x86 line. The Niagara line of processes have had hypervisors as an integral part of the system since the first generation T-1 processors in the T-1000 servers.
Don't even know wtf you are talking about as far as "architectural design does not simplify operations and make IT more agile." Solaris supports all modern technologies commonly found in a data center.
Re: (Score:2)
Right there, you're tacitly admitting that there's nothing worthwhile about Oracle's SPARC hardware. If it was faster, or cheaper, or one of the above for some subset of tasks, there would be a huge number of reasons you might want to run Linux on them. Linux is sufficiently cross-platform that moving workloads from x64 to SPARC would be reasonably easy, while porting eve
Re: (Score:2)
What I was trying to do was phrase things in a way to avoid triggering the reactionary responses of people who are emotionally invested in Linux and are completely incapable of hearing any suggestion that there might be some system somewhere that runs better on something other than Linux. Apparently I failed.
Here's the biggest reason to run Solaris rather than Linux on SPARC: The builder of the hardware also writes the OS. This means a whole bunch of things, for example you'll never encounter a situation wh
Re: (Score:2)
Solaris runs on x86 indeed, but the pricing will kill you, and is such that it makes SPARC servers nearly free if you need Solaris.
I actually know Solaris fairly well, and it has its strong points, but I can't say the same for SPARC hardware. Oracle and IBM have fared somewhat better than HP (who are dragging out the slow death of the Itanium) but the downfall of proprietary platforms and vendor lock-in, is on the horizon. They just can't compete. SPARC machines used to dominate the top ranks of the TOP5
Re: (Score:2)
Fastest CPU? (Score:2)