Five Years After the Sun Merger, Oracle Says It's Fully Committed To SPARC 190
jfruh (300774) writes "Sun Microsystems vanished into Oracle's maw five years ago this month, and you could be forgiven for thinking that some iconic Sun products, like SPARC chips, had been cast aside in the merger. But Oracle claims that the SPARC roadmap is moving forward more quickly than it did under Sun, and while the number of SPARC systems sold has dropped dramatically (from 66,000 in Q1 '03 to 7,000 in Q1 '14), the systems that are being sold are fully customized and much more profitable for the company."
I'd love to buy some sparc hardware (Score:5, Insightful)
If it wasn't for that the price of the hardware can often be close to ten times higher than the equivalent x86 machine.
Re:I'd love to buy some sparc hardware (Score:4, Informative)
It should be: around the time of the acquisition the price performance ratio finally got back to where it was with the first SPARCs: ten times the price, 100 times the performance .
A 3U 4-socket T5 machine had about 128 full hardware threads (really: cores) the last time I looked seriously at it. The performance was a bit less than a 32-socket, 4-core-per -socket M9000, the machines I mostly worked with. In those days, I was a capacity planner and performance engineer at Sun Canada.
A lot, but not everything, is still available open-source from SPARC International.
--dave
Re:I'd love to buy some sparc hardware (Score:4, Informative)
A 3U 4-socket T5 machine had about 128 full hardware threads (really: cores) the last time I looked seriously at it. The performance was a bit less than a 32-socket, 4-core-per -socket M9000, the machines I mostly worked with. In those days, I was a capacity planner and performance engineer at Sun Canada.
Okay, but it's not a one-for-one comparison. A hardware thread in one of those machines lies somewhere between an x-86 "core" and a GPU "core" in capability.
Granted, they were powerful machines. But a 128-thread SPARC machine has nowhere near the capability of a modern x86 machine with 128 cores.
Re: (Score:3)
Just for comparison, that was about a 3U machine.
You can fit 64 cores and 512GB (1TB if you're rich) into 1U with commodity x86 servers. Last time I checked, power draw is about half the lifetime cost, so the SPARC servers would have to be awfully good on power draw/performance.
On the plus side, the big T5 servers have larger system images than the cheap commodity x86 ones. So if you want to keep a fuckload of data in RAM, they could be worthwhile.
Re: (Score:3)
If it wasn't for that the price of the hardware can often be close to ten times higher than the equivalent x86 machine.
At least for certain definitions of "equivalent" ...
Not so. (Score:5, Informative)
If you examine the top two best performing database platforms [tpc.org] (as benchmarked by TPC-C score) you will discover that they are both sold by Oracle, and that the SPARC version has both higher performance and a lower cost per transaction than the x86-64 version.
You might find this quote [theregister.co.uk] to be particularly interesting:
Re: (Score:3)
From the table, it looks like the Oracle is the fastest, but also the highest price/tpmC, while the Dell is the cheapest.
Re: (Score:2)
The Dell server is actually running a Sybase product, which is 98% slower than the SPARC benchmark. It is the newest entry on the list.
If you want an inexpensive database, you might look at Oracle XE [oracle.com], which is free. However, it has some rather tight constraints and limitations, and it only runs on x86.
Re: (Score:2)
The Dell server is actually running a Sybase product, which is 98% slower than the SPARC benchmark. It is the newest entry on the list.
If you want an inexpensive database, you might look at Oracle XE [oracle.com], which is free. However, it has some rather tight constraints and limitations, and it only runs on x86.
"Free". Unless you have plans to upgrade to full blown Oracle at some point in the future, I see no reason why you shouldn't be using Postgres instead.
Re:Not so. (Score:4, Insightful)
If I want Oracle PL/SQL in Postgres, I have to purchase EnterpriseDB [enterprisedb.com]. If you can get EnterpriseDB to give away the "deep Oracle compatibility" for free, many Oracle installations might switch. Let me know how that works out for you.
I'd also like to see PostgreSQL in the TPC-C top ten. That's a lot of work, and for people who need scalability, they don't have time to wait.
Re: (Score:2)
I don't care about PL/SQL or deep Oracle compatibility. As I said, if you plan on upgrading to full Oracle later, go for the "free" Oracle product.
Re: (Score:2)
I work on nothing but open-source based PostgreSQL servers for living, effecively competing against EnterpriseDB. All of the Oracle migrations I've done convert their PL/SQL code to PostgreSQL's PL/pgSQL instead. There's always some amount of application and process refactoring when you're moving to another database; this gets wrapped into that. I do a few training classes each year on the quirks of PL/pgSQL, and most companies don't do anything so complicated in there that they can't move things over.
It
Re: (Score:3)
The question is- is that because sparc is better, because Oracle optimized for Sparc like mad, or because they purposely degraded the performance on x86?
Re: (Score:2)
While either scenario could be true, the x86 score is triple the performance of the next best-performing platform (a DB2 database from IBM). Oracle rules these benchmarks, even after failing to submit a new x86 system for over two years.
Re: (Score:3)
How about look at TPC-H [tpc.org], where in the 10TB test the T5-4 gets beat in performance by the DL580 G8 and the x64 system is half the cost per transaction? The number of shops that need more than 120 threads and 3TB of ram is vanishingly small.
Re: (Score:2)
The published dates for the tests were less than 5 months apart, and the reason they used SQL 2014 is that it's what their customers are actually using, fewer shops are picking Oracle due to their licensing stupidity.
Re: (Score:2)
You might find this quote to be particularly interesting:
"I am going to make a promise to you," [Larry] Ellison said. "By this time next year, that Sparc microprocessor will run the Oracle database faster than anything on the planet."
So, how many nops are going into the other versions?
IBM should submit some (Score:2)
It's up to the vendor to assemble and benchmark a database platform for TPC.
Software licensing for most database servers explicitly prohibit benchmarking, meaning all scores must be released by the vendor. IBM likely has seen fit not to, for reasons of their own.
Re: (Score:2)
I agree that free software is a good thing, and so does Oracle, as they give away their RedHat clone for free.
Oracle gives away PostgreSQL, MySQL, SQLLite, and several other databases in their clone, and will support them if that is important to you.
Some applications require scalability, availability, or other features that are beyond the realm of the free databases. In those cases, Oracle database XE is free, standard edition is under $10k, and enterprise is available where performance outweighs cost.
I who
Re: (Score:2)
A free product is still a free product if the support isn't free.
Re: (Score:3)
Sun was trying to sell hardware, I guess we all know how well that went. My guess is you said exactly the same before Oracle bought them. And you don't want to compete with Intel on economics of scale when the vast volume of x86 servers would never migrate to sparc, it'd be like a swimming contest wearing a lead vest. Oracle has quite rightly assumed that if they want to sell sparc hardware they have to create the market by making it the most cost efficient way to run Oracle the database. For anything else
So how many Sparc Systems does Oracle Run? (Score:5, Interesting)
While reading TFA, my big question was if the Sparc has been improved so much, is Oracle using it in their systems?
According to Wikipedia, Oracle has 122k employees; how many of them are running Sparc systems, how many of their internal servers are Sparcs? For a corporation of this size, I would expect, in three months, for them to consume a lot more than the 7k systems that were shipped in the latest quarter.
When I was at IBM, the company was very proud to be its own best customer; is that true for Oracle?
myke
Re: (Score:3)
IBM no longer sells desktop and they're getting out of the server market as well, so I think they'll be Lenovo's best customer for the foreseeable future.
As for Oracle, my guess is that these are Big Beefy Machines(tm) used as replacements for the IBM mainframes (which IBM still owns). They probably do use some in their back-end gear, but don't forget that Oracle also owns Oracle Linux and they have their own line of x86 hardware. That's more likely what they have most of.
Re: (Score:3)
They are getting out of the X86 server market, not the entire thing. They will still be selling power.
Re: (Score:2)
Yeah, I corrected myself later on and didn't go back and change the original statement.
Re: (Score:2)
Given the fact that they outsourced Power chip production you have to wonder how long that will be true. Then again HP outsourced their processor development to Intel over a decade ago and they still sell Itanium, so it might be a long time before they give up that gravy train.
Re: (Score:2)
and by "sold" you actually mean "given to with a fat stack of cash". IBM is paying GlobalFoundries to take the semiconductor fab business from IBM.
Re: (Score:2)
I believe the GP may have been talking about the Power/POWER based line of servers & workstations, as well as their other iron. But you're right about Oracle - their homegrown systems are Lintel hardware (their own distro).
Only thing I've wondered - if they're that committed to SPARC, why don't they have it populate more of their line, instead of Xeons/Opterons? They could port OEL to it, in addition to Solaris, and customers would be locked on that - couldn't move those to HP/UX or RHEL or Centos o
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
....and refusing to share the secret sauce.
Is that like Colonel Saunder's secret recipe?
Re: (Score:3)
IBM was (is?) split into different fiefdoms and the PC division hated OS/2 and loved Windows and at the end they had enough pull to force the cancellation of OS/2-PPC when MS was going to refuse any special pricing if IBM continued with OS/2 development. Gartner also didn't particularly like OS/2 as well.
Re: (Score:2)
Re: (Score:2)
(IBM) Employees now are issued laptops with a rebranded version of RHEL installed.
Well, now that Apple and IBM have a cooperation deal . . . maybe we will see IBMers showing up to meetings with Apple laptops . . . ?
Re: (Score:2)
"Employees now are issued laptops with a rebranded version of RHEL installed."
Why not SCO Linux? Given that SCO and IBM are such close business partners. Typical example:
http://prwire.com.au/pr/4550/i... [prwire.com.au]
"IBM DB2 Version 8.1 Certified on SCO Linux 4.0 ..."
Re: (Score:2)
Some time ago when I was working for IBM, I did have a POWER workstation (back in the days they were called the Risc System 6000). Our entire department had them.
Re: (Score:2)
Re:So how many Sparc Systems does Oracle Run? (Score:5, Informative)
Fore disclaimer: I'm employed by Oracle.
My duties at Oracle have always been developing Mac and (more recently) iOS Apps. So I end up using Apple hardware. That aside, all developers have a remote unix account for development. We host server instances of the particular product we develop for, and have other such servers for hosting a number of dev tools. Many of those servers are hosted on "Sunacle" (my term) hardware.
In the Mtl office, we also have a bunch of Sun stations in various places like some local server rooms and demo/training rooms.
Sun is *everywhere* around me. Oracle has a huge investment in Sun (both in hardware optimisation and Java) which led to the acquisition 5 years ago. Was more of an investment rescue than a growth acquisition, if you ask me. BUT I DO NOT KNOW for sure if my point of view & assessment is what really led to the acquisition. Developers are not privy to such details.
Re: (Score:2)
Have you considered typing that question into your Ask toolbar?
Unfortunately... (Score:4, Insightful)
Re: (Score:2, Insightful)
Why would you use virtualization in such an environment? Not trying to be argumentative, but it doesn't seem like virtualizing a bunch of database servers would be that big a win. So many people post here about virtualization that I'd like to know what they find so useful about it.
I personally just don't like the concept of adding a layer between the OS and the hardware. The OS is supposed to handle running different programs and providing the environment they need to run. If it's not, fix the OS, don't
Re:Unfortunately... (Score:4, Informative)
Re: (Score:2)
Don't forget replication. A company that makes 15 million a year would never have considered 4 hour DR but with the low cost of hardware and VMs you can easily do this for a very small amount of money and time.
Re: (Score:2)
I agree, a SAN, or zfs does mimic many of the advantages.
Re: (Score:3)
Except all the ones that do. You should get out more.
Re: (Score:2)
Zfs filesystems do the same, vmware is easier to use thats all. Sparc is a niche market for large corporations. (Banks, financial services, payrolls), thats why is expensive. There is no way a large corporation will use a cheap commodity hardware and vmware for mission critical operations.
That may be true for corporate dinosaurs, (big old companies that started back when having a corporate song [arstechnica.com] sounded like a good idea) but most corporations that got started after the 90s have only ever lived on commodity Intel/AMD (in some cases even ARM) hardware.
Re: (Score:2)
Even corporate dinosaurs that have legacy mainframe operations that go back decades are moving to VM clustering because of the inherent advantages. When I worked for one of these dinosaurs last year, I was walking through the mainframe operations group and someone had a thing printed on the side of her cube extolling that it would take 400 x86 servers to equal their S390. Someone from the VM engineering team had taken a sharpie to it and crossed out the "400 x86 servers" and written on "10 VMware hosts".
I
Re:Unfortunately... (Score:5, Interesting)
I can sum it up in one phrase: No. Hardware. Downtime. Ever.
VMWare's solution enables you to move production servers at will without ever halting execution. Any hardware upgrade/replacement will have zero downtime. Even a hardware failure can be automatically migrated away from before it takes down the server and fixed without any down time.
Re: (Score:2)
I get that, but you can do process migration without virtualization: http://criu.org/Main_Page [criu.org]
Is it just because that's experimental still?
Re: (Score:2)
It's experimental now, but it won't always be. Are you saying that you won't use it because it's experimental now, or you won't ever use it because you think there is something fundamentally wrong with the approach?
Re: (Score:2)
Well, it can but we've had a few underlying hardware issues make our servers unavailable. From san hiccups which flip systems into read-only mode for /var to a recent issue where a VM was configured as Fault Tolerant and it was VMotioned to a different system and connectivity was lost. Turns out the MAC changed, maybe during a VMWare update, and apparently because the VMs were FT, the MAC didn't change so became unavailable until we powered down the VM to recover.
There have been enough issues that we won't
Re: (Score:2)
Even a hardware failure can be automatically migrated away from before it takes down the server and fixed without any down time.
The history of automatic fail-over software says that when you add some, in addition to the old issues you now have bugs in the fail-over software as a new problem.
Re: (Score:3)
That's nice, and for those of us who want to spend less than $1M per year per box there's VMWare which provides nearly the same uptime at a fraction of the cost, and we don't have to put up with IBM. Commoditization is a good thing for the customer.
Re: (Score:2)
I guess the parent means vms that run linux on sparc.
Solaris on sparcs has VM support build in, they call it a 'zone' that is an extreme close to haedware virtualization/chroot approach.
VMs help if you regularily need 'boxes' that you simply throw away after they got used a few days and can simply be re-instantiated when you need them again. That is usefull in software development especially testing.
However we are talking about big iron here. When a Sparc mainframe has 256 cores and 1TB Ram it might be spli
Re: (Score:2)
Uh, DL580 G8, 120 threads and 3TB of ram. Heck, my VMWare hosts are 40 threads and 384GB of RAM, we run everything except Oracle DB and OBIEE on them, and only then due to licensing idiocy, not capability or reliability.
Re: (Score:2)
What do you mean with 'threads'? Cores? :)
Don't get your point. But seems to be a nice box
Re: (Score:2)
60 physical cores + 60 hyperthreading units, kind of like how all the current gen SPARC processors have 8 thread pipelines but two execution units per core. The T5-4 has 64 cores and 128 execution units and the performance is actually pretty close to the DL580 Gen8 so it's not a horrible way to compare systems. Oracle systems go a bit bigger with the T5-8, but my point is that you can go nearly as big with x64, and VMWare allows you to create essentially arbitrary sized containers from the hardware, just li
Re: (Score:2)
Then you suck at tuning, it's trivial to get 90% throughput VM versus physical, and with a bit of effort you can achieve 95+% (VMWare actually managed to get over 100% on some Java tests due to scaling issues with the JVM, running lots of clustered smaller instances on multiple machines was faster than running bigger instances or lots of small instances on one machine).
My suggestion to Oracle: SPARC everywhere... (Score:4, Interesting)
My suggestion to Oracle: Get SPARC's marketshare up. This might take some doing, but long term, expanding the ecosystem is a good way to keep revenue coming in, where customers buy new machines to upgrade, as opposed to "upgrading" to commodity x86 hardware.
This would require some work on the whole stack from the CPU on up to applications. For example, getting Solaris LDOMs and domains to work with SCVMM or the enterprise admin tool of choice. Another would be getting Linux applications to work on Solaris with low to minimal porting necessary. IBM did this with AIX starting at 5L (where it took a code recompile, but little else.)
As I mentioned before, Oracle has some pretty nice technologies which can shake up the market. SPARC servers have Infiniband, so if Oracle does some work with the hypervisor to allow one machine to access another box's disks via Infiniband, add redundancy (on both drives and nodes), this would completely get rid of a need for a SAN backend. Need more storage? Just add more drives to one of the machines, or add another node to the cluster, similar to how Isilons are updated. ZFS is also a crown jewel, and can be used for a lot of things as well, especially backend deduplication.
I hope Oracle can reinvent itself. They have a lot of core technologies that they could use to eke out a definite niche in the enterprise. Combine that with the fact that SPARC and Solaris are mature technologies, and Oracle can bring to the table pretty decent security.
Re: (Score:2, Insightful)
That's not how Oracle makes money. They buy popular but less profitable companies, and then jack up the prices on their product until everyone finally migrates to other systems. Once they've driven away all the customers of the acquired company, they buy another popular but unprofitable company and repeat.
Re: (Score:3)
That's not the only way Oracle makes money.
They also get companies to sign unreasonable contracts, then six months later 'hire away' the deal maker for 5-10x previous salary for a zero responsibility marketing job that lasts a few years.
If you ever see that pattern on a resume _run away_. Not only is the person crooked, they can't manage money. The job should have left them set for life. Some are so greedy they try to leverage the salary history.
Re: (Score:2)
Re: (Score:3, Informative)
AIX 5L+? Minimal porting? You've very obviously never actually done it.
The total extent of IBM's efforts with AIX 5L was to put RPM 3.0.3 on their systems and build a few RPMs. The underlying source base for your RPM better support AIX or you're in for a good deal of fun. And you know what? Pretty much everybody dropped AIX support years ago for, I might add, very good reasons. AIX is a Unix, but a seriously weird one. Oh, and by the way, can you guess the version of RPM shipped with the latest AIX? Clue: i
Re: (Score:2)
Re: (Score:3)
I do agree that AIX does stand for "Alien Interpretation of UNIX", but even though it is squirrely, if an application runs on it, it runs well.
I am not disagreeing with the fact that AIX and Solaris are bit players. However, I would say that one problem is that both Oracle and IBM at best are focused on retaining existing customers. Neither have any marketing focus on getting people from VMWare and OpenStack onto their platforms. And without expanding the market, just as the parent stated if the market i
Re: (Score:2)
I also had quite a few issues with one of my two HP-UX systems (HP-UX 11iv3 on Itanium, the HP-UX 11 on RISC I don't think anyone ever used).
"each system is more profitable for Oracle" (Score:3, Insightful)
I'm glad that each system now makes more money for Oracle, I knew there was a reason for buying Sun/Oracle gear, it makes them richer.
Just for a moment I thought there might be a reason *for me*.
Re: (Score:2, Interesting)
I work for a company that was one of Sun's top 5 customers [ergo, "Anonymous"]. When Oracle took over, we were greeted with the elimination of most of our bulk discounts and an admonition to no longer deal with VARs. Since then, we've begun an aggressive Linux implementation program. The purchase of new Sun-branded hardware, in my rather small working group alone, has gone from several hundred servers a year to zero. We have pallets of decommissioned T-series machines in storage awaiting a trip to the sc
Re: (Score:2)
If you ever thought Oracle is not a behemoth exclusively after your money, then you probably never had to really deal with them...
lot of talk (Score:2)
Lot of talk about number of cores, and cache sizes, but what is the actual performance compared to intel's chips ?
Re: (Score:2)
Lower performance with higher power consumption?
Re: (Score:2)
Looking at spec.org jEnterprise2010 scores:
http://spec.org/jEnterprise201... [spec.org]
http://spec.org/jEnterprise201... [spec.org]
A T5-2 gets a jEnterprise2010 score of 17k, an X4-2 11k (with half the memory and Oracle Broken Linux 5.9, why not 6.*?).
The sparc has a list price of ~68k USD. Not sure what a two socket Oracle intel box costs; maybe 15k or so?
sparc; 4 usd/score
intel: 1.36 usd/score
Sparc was nice once, but that was ever so long ago..
Re: (Score:2)
SPARC really excels at highly multithreaded and highly multi-process workloads. Its sequential throughput leaves a lot to be desired but it really shines on some enterprise applications.
Hmmmm... I forget... (Score:2)
Is "totally committed" the same as "100% committed" in business speak? I think they both mean "We'll do whatever it takes to keep the most money rolling in." But that's probably cynical, as the operators of corporations have nothing but the kindest of motives towards their customers in their generous hearts.
Less than 110% (Score:2)
Re: (Score:3)
110% = 25% mon-thr + 10% fri
Re: (Score:2)
You have to be kind to your customers, otherwise they move too much when you rape them.
Well, that's quaint... (Score:2)
What do Oracle Customers Say? (Score:4, Interesting)
I worked at an Sun Micosystems shop. We bought thousands of their servers yearly and these wren't just cheap system, but the big E-class stuff for $500K-$3M each. The people were good to work with, the hardware lasted just a little longer than we wanted, and Sun was a nice company for the F/LOSS world.
Then IBM offered a better golf deal to the CxO at that place and we were directed overnight to buy IBM whenever possible. The P-class stuff was cheaper than Sun's and AIX wasn't hard to use - we ran Sun, IBM, HP, and a few other systems - not a big deal.
After a year, Sun came back with new architectures that added many more cores for next to nothing extra power. We went through a huge modernization effort to free up physical space in all our data centers and deployed virtualized servers as a default. It was fairly routine to swap 1 physical box for 10-20 older boxes. Nice.
Then Oracle bought Sun and started the marketing takeover. Engineers know what I'm saying (VMware/EMC are similar). Then Oracle started behaving badly in the F/LOSS world, killed a few projects and started to stink up a few other projects.
Never pay Oracle for anything except a DBMS - Oracle. Don't get consulting, and run, run, run away from their enterprise software stuff. Anyone who has been through 2-3 yrs of attempted deployments for these white elephants knows why. You will be sold the impossible and it will never be completed. At $300/hr per consultant, they will bleed your budget until you can find a scapegoat to fire, thus saving your own career.
For us, the writing was clear - only buy Oracle HW when absolutely necessary and reduce our dependence on their DBMS to about 10% of our DBs. Go with Linux and x86 hardware whenever possible and use postgres for the DB unless really needed so there was real competition.
What do other customers say?
But we're not committed to Oracle (Score:2)
Oracle Says It's Fully Committed To SPARC (Score:2)
RIP SPARC.
Consistent with a dying platform (Score:5, Interesting)
A more telling stat was that in Q1 2003, Sun shipped 66,000 Sparc units, most of them Sun Fire servers, the commodity line. In Q3 of 2014, that number was down to no more than 7,000 units in the quarter. But he notes that while Oracle's unit sales are down, the devices it sells are very high-end and are fully configured and integrated with compute, storage, networking and software completely integrated.
That isn't a refutation of the claim that Sparc is dying, it's just an explanation of how it happens.
Sparc users are the same as any other group, the exodus starts with the fringe and then moves to the core. Casual low-profit customers found it easy to switch platforms so left a long time ago. The big high profit customers have high loyalty and massive sunk costs, it's hard for them to switch platforms so they'll be the last to go. If Sparc is dying then that's exactly the pattern I'd expect it to follow.
Reputation (Score:3)
Oracle says a lot of things.
They inflicted Fusion on my employer, and not a single claim about it has turned out to be true.
Actual Solaris Sysadmin Here - Here's the story (Score:5, Informative)
Solaris/SPARC is still going strong in large companies. One of the greatest advantages it has is that Oracle creates and supports the operating system, and Oracle creates and supports the hardware. (If you're running an Oracle database or some other piece of software, then that's an additional component that they create and support.) What this means is that if I'm having a problem, mundane or esoteric, I can go to one vendor and say, "Fix it." There isn't any bickering about what company's problem it is, and who manufactured my RAM, or any other the other silliness that crops up in vendor support. Large companies value this (as do us sysadmins). That also means they can do some very cool software tricks (which I'll mention a few here below).
The decreasing unit shipments is just as much a sign of virtualization as anything. Right now, I'm looking at an older T5240, with two eight-core CPUs which presents itself as having 128 virtual CPUs (execution engines or thread engines), and 64gb of RAM. This is by no means the biggest box on the floor. We carve these up into smaller systems using either Solaris Zones, or LDOMs. That's two different methods of virtualization with two different goals.
I did something great with an LDOM last week. I took a virtual server that was on the box and migrated the entire operating system and all the applications over to another LDOM... WHILE IT WAS STILL RUNNING. Aside from a quick (1 second) pause, the applications on the server had no idea that it just migrated to another piece of hardware while it continued to run. Slick! The original server had a failing DIMM. No worries, though even aside from ECC, the operating system automatically mapped out which parts of the DIMM were defective and retired the pages of memory so that they weren't constantly being exercised. Linux does all that... right? No?
Someone else, above, said, "I don't think you can have a zfs system fail and move it to different hardware like you can with vmware...". Nevermind that we can migrate a running operating system and application to another piece of hardware and keep it running. Yes, of course if you have a hardware fault, you can bring it back up on another machine. The virtualization with Solaris is quite capable.
In the environment of a large company where we're competing against Linux on the low-cost end of things, Solaris/SPARC is not only holding its own, but actually beating our Linux cloud counterparts in the costs of a virtualized OS/hardware. (I should ask my boss if we can publish a paper on this, because it is rather impressive.)
On the high end of things, we completely dominate. We generally use a T5-4 for our internal cloud (which really isn't the biggest Oracle server out there). It has 64 cores, presenting 512 execution threads to the scheduler. RAM goes up to 2TB. If someone starts out on a tiny box with only one CPU and 4gb of RAM, we can scale them all the way up to the top by increasing their virtualization settings. No migrating to different or unusual hardware. If an application team can't scale their code horizontally (hey, it happens), they can go way vertical in this configuration. We haven't had a need yet for an M6-32 (32tb of RAM, and 32 of the 12-core CPUs (3072 execution threads or "virtual cpus"). We have Linux surrounded (on the low-cost side and the high-performance side) in a large enterprise environment, and that's why Solaris is still there.
Now, I'm not an Oracle salesperson. But if Slashdot ever did an AMA with an Oracle sales engineer, I think my fellow Linux admins would be particular impressed on how far ahead Oracle/SPARC is in a number of key areas.
Re: (Score:2)
> Also, yes, various Linux virtualization technologies you can hot-migrate running systems between software VMs within the same chassis (and warm migrate with downtime between hardware chassis).
In this case, it was a live migration over two different chassis.
> But the price/perf/features just don't add up in the modern world versus commodity hardware and open source software, except...
In the existing case with a large corporate environment with a Linux/x86 cloud and a Solaris/SPARC cloud, Oracle wins
Re: (Score:2)
> Why bother? Just shut down that server, replace the memory, restart. If your application can't handle a brief downtime for one of your servers, there is something wrong with the application, and no OS magic can fix that for you.
You know, it is kind of funny. Person A will argue, "See? Linux has all of the cool features of an enterprise class operating system. What makes Solaris on SPARC so special?" When you point out just a fraction of the things that Linux doesn't do, person B will jump in and claim,
Re: (Score:2)
> SunOS/Solaris started out lean; when it got bloated, people like me jumped ship. Linux started out lean and it is getting bloated; when it is getting too bloated, I will jump ship again.
FWIW, Linus said that Linux is bloated. That was at LinuxCon in 2009. I don't think that's much of a factor. I'll agree with you that Linux has plenty of life left in it.
Re: (Score:2)
Wow. It took them that long to put in a handler for CE and UE events? I see. What we're talking about goes beyond that first level of handling those events.
Re: (Score:3)
> Wow. How impressive. Oh wait, Linux has had EDAC since 2006. But you keep paying your millions to Oracle. I'm sure its worth it.
Actually, this might be worth an illustration. It was a long time back, so I'm sure I've forgotten a few details, but I'll give you the big picture.
Around 2000, Sun Microsystems had a problem with the L2 cache on their 400mhz CPUs. It seems that IBM misrepresented the error rate on the chips, and they were having bit errors that were much higher than specified. Because of what
Re:Actual Solaris Sysadmin Here - Here's the story (Score:5, Interesting)
I remember the first time I had a real hardware error on a Dell system running Linux. Straightforward enough, called out DIMM1. So I called Dell. They said "Oh, that doesn't necessarily mean a memory error. The way the PCI bus works that error could be on the bus itself, in the memory, or in the card in the first PCI slot. There's no way to tell".
Seriously? No way to tell what "Error in DIMM 1" means? That's what the guy insisted. His solution? Turn the computer off and reboot. If it crashes again, call him back.
This was on a Production database. No way was I going to just power off/on and wait for a follow up crash. I was used to sending Sun explorers and getting exact part numbers back for failures. If Dell couldn't do that, why were were playing this game?
Dell finally agreed to send a technician with all three parts so he could diagnose and we could solve it with one downtime instead of several. But as a long time Solaris guy, I was totally disgusted.
Sure, for edge servers, startups, small things, you can get away with that. But for business critical in Enterprise? I want better support from my vendor than "reboot and let us know if crashes again".
Intel (Score:2)
Intel insisted they were "committed" to Itanium for a lot of years after that horse died, too.
Re: (Score:3)
Re: (Score:2)
much less. Maybe the ability to run it in a couple of 3U boxes: I'd have to spreadsheet it to see.
Re:Why SPARC? (Score:4, Informative)
From a general standpoint, very little - which is why their numbers have dropped from 66k to 7k in 11 years. They are apparently used to build systems still compatible w/ legacy Solaris systems, which is what enables their high margins. Otherwise, more than 64-bit, Intel's multi-core architecture, and the fact that it is several process nodes ahead of the SPARC, gives it a big advantage at the same price point, not to mention its support for several more modern OSs, such as Linux, BSDs, Windows Server 2012, and more. If you don't have a legacy SPARC establishment, there's no reason to go that route.
As Intel found out w/ Itanium, the traditional disadvantages of CISC - wrt not only VLIW, but RISC as well - are obliterated: Merced originally resulted in only a 10% savings in die size - certainly not worth the complexity in the compilers and other costs incurred in building that platform. And once Intel tossed more cores into a CPU to scale up its performance, overtaking any other RISC CPU at the same price was no longer an issue. Especially since every OS for it - Windows, Linux, *BSD - support it
Re: (Score:2)
Re: (Score:2)
The bean countress was right, maybe for the wrong reasons. Reason HP/UX is a bad idea on x86 is that if a customer was gonna move to x86, they already have plenty of options - Linux, *BSD, Solaris, OpenIndiana, et al. And I didn't even count Windows Server.
If one does an ROI, it would be a tall order to recoup the costs of an x86 port. If a customer has to replace their PA-RISC/Itanic towers w/ x86, they'll probably try to go all the way, and replace HP/UX w/ Linux. Since HP apparently already suppor
Re: (Score:2)
Re: (Score:2)
An "Artisan Processor"? (Score:2)
Or maybe "Heritage"?