Excellent news! This article cites real, unassailable numbers-- much of them in *dollars*. There is ample statistical basis to draw many, many well founded conclusions. These conclusions will affect many types of business, economic models, political systems, artistic expressions, and maybe even sports. I would humbly suggest that every single one of those conclusions bodes well for the careers of (stereo)typical readers of this site.
ZFS already supports flash devices for caches. For read caching (L2 ARC), you can create striped cache volumes. You get better speed that way, and if one of the devices fails, ZFS knows it and just goes straight to the main storage volume (the one being cached). Meanwhile, the other drive continues. For write caching (ZIL), since the data is "worth" more, you can create a mirror of flash devices. The benefit of the ZIL is realized even if the cache is small, but unfortunately SSD write speed can be worse than writing to regular drives (see below about SLC drives).
The best theoretical configuration would be, at a minimum: - two small (and fast) SLC devices, mirrored and used for write caching - two large(r) MLC devices, striped and used for read caching - a redundant array of inexpensive drives (someone should come up with a catchy term for that), of huge capacity but otherwise slow (5400 rpm)
In place of the SLC drives, there are even more expensive (but higher performing) options, such as a bank of volatile RAM with a battery backup, and an SSD that the RAM contents get copied to in case of a power loss. These exist, and really work. The theory of a pyramid of caching; with "slower and cheaper" at the base and "faster and pricier" towards the top really has been shown to work.
ZFS can do all of this right now, and continuing a little off topic... can also do compressed incremental volume snapshots sent into the cloud:)
Yeah, I do a lot of work with ZFS. All of this stuff really works.
Maybe not science fiction, per se, to some-- but I think it counts. And don't worry, the book is an order of magnitude better than the crappy animated movie which I will not spell out:)
Don't make the mistake of confusing "the world I want" with "the world that is". Facebook is a private enterprise, it is free, and it is deeply flawed.
I wish I could park my fancy convertible outside the liquor store with the engine running... and it still be there when I get back. Alas, I must suffer through the world that is. And someone is going to steal my car if I make it too easy.
Who thinks Facebook is private? The whole point is to *not* be private, right? Otherwise... what is the point of Facebook?
If the FBI was going to start monitoring encrypted email, VPNs, and other things where you are *trying* to be private, I would be concerned (yes, I know-- whole 'nuther can o' worms). But Facebook? You are giving the info away as a user, that is the purpose of having a Facebook account.
I just got an A77, and I can't understand how anyone could knock the image quality. It has 24 mpixel, I am not impressed with 1:1 blowups detailing every errant pixel.
Having said that, the A77 is probably not what the poster is looking for, and if the NEX7 is half the camera the A77 is (and it is probably more like 90%, which makes me *almost* regret getting the A77), you won't regret it. Good luck finding one until April:)
Sometimes, you don't want to "look" like a photographer. A high quality EVIL camera allows you to be a little more inconspicuous and still get away with great shots.
If you can find one, get the NEX7, and pick a E mount lens that best suits your needs. Then, get an L to E adapter, and you can use 100's of Leicia L mount lenses. Most will only work in manual focus mode, but the peaking display makes focusing easy and accurate.
I just got an Alpha77, which is probably not what you are looking for. I love it, though, and the NEX7 shares a lot of the same technology and features.
When this whole CarrierIQ thing got started, I thought this was simply a diagnostic tool, yet conspiracy theorists were going to jump all over the "...but they could" aspects of the system. I also thought it was a shame, since the carriers and manufacturers ought to be able to monitor the system so they can improve it.
Your post has made me re-think all of my notions. I don't believe any nefarious purpose was afoot-- this was a tool intended to diagnose infrastructure and device performance. However, installing it as a rootkit is a bad call. It provides a vehicle for malware, and a description of its operation-- however technically accurate it migh be-- touches too many evil buzzwords. Such a tool, while useful, is eventually too easy to turn into a PR nightmare (obviously). Then throw the malware hijacking aspect in for good measure.
Verizon does it right, at least as we see it now. Those vans do a great job of real-world testing, where their test equipment is gathering the same metrics as CarrierIQ's software, but with test data nobody will whine about.
But...
...couldn't those vans also spy on every packet going through the cell they were testing? I'm not suggesting they are, or if that would be on any use. But they certainly have the necessary equipment in those special vans. Paint the vans black... and... wow, I don't want to think about it.
Pick your country and protocol. Works with the BBC, Hulu, Vimeo, Pandora, Netflix, probably just about everything else. Of course, you have to pick a US proxy for Hulu, a British proxy for the BBC, etc.
I could not agree more. I have used many of the GEOM features, they all work perfectly. The partitioning scheme, in particular GPART and GLABEL, make working with USB and other hot-swappable devices much easier and less error prone.
The FreeBSD motto is "the power to serve", linux laptop users should keep that in mind.
I had not seen that article yet, thanks. However, it still misses the point. ZFS is *not* intended for 4GB workstations with single hard drives!
I'd like to see them benchmark a ZFS setup with 3 or 4 RAIDz pools concatenated into a 10 to 20 TB partition spanning 10 to 15 physical drives, with an MLC L2ARC. Throw 48GB of ram and an 8 core processor at it as well. Next, properly tune the InnoDB and logging partitions. Then, benchmark MySQL. Finally, try a large update/join spanning many tables, using foreign keys and stored procedures-- and: -- pull the power plug, see if there is any data corruption -- do a ZFS snapshot, live, see if the benchmark notices the difference -- add more RAIDz pools to the live file system
Sorry to hype on ZFS so much, but that is where my expertise is. Their benchmark was testing a scenario that nobody would be encouraged to ever actually use in the real world. It makes me seriously doubt the usefulness of their other benchmarks (like HURD) when they make such an obvious (to me) mistake as to omit these important details.
Phoronix has a history of questionable choices for their benchmark setups. Hardware, versions, and tuning are... cleverly chosen, almost as if there was a preconceived agenda with inevitable results. Not that there is one-- just like it seems like there is. And so colorfully presented! I remember when they tested ZFS on an i386 version of FreeBSD on a 1G laptop! Others have also noticed this Phoronix phenomenon:
The whole point of Hurd, at least right now, is tangential to benchmarks. Nothing wrong with testing, of course, but I think the results should not be used for any long term planning. Nobody is planning on launching a business running on Hurd servers... yet.
Slashdot Mirror
Excellent news! This article cites real, unassailable numbers-- much of them in *dollars*. There is ample statistical basis to draw many, many well founded conclusions. These conclusions will affect many types of business, economic models, political systems, artistic expressions, and maybe even sports. I would humbly suggest that every single one of those conclusions bodes well for the careers of (stereo)typical readers of this site.
Enjoy!
ZFS already supports flash devices for caches. For read caching (L2 ARC), you can create striped cache volumes. You get better speed that way, and if one of the devices fails, ZFS knows it and just goes straight to the main storage volume (the one being cached). Meanwhile, the other drive continues. For write caching (ZIL), since the data is "worth" more, you can create a mirror of flash devices. The benefit of the ZIL is realized even if the cache is small, but unfortunately SSD write speed can be worse than writing to regular drives (see below about SLC drives).
The best theoretical configuration would be, at a minimum:
- two small (and fast) SLC devices, mirrored and used for write caching
- two large(r) MLC devices, striped and used for read caching
- a redundant array of inexpensive drives (someone should come up with a catchy term for that), of huge capacity but otherwise slow (5400 rpm)
In place of the SLC drives, there are even more expensive (but higher performing) options, such as a bank of volatile RAM with a battery backup, and an SSD that the RAM contents get copied to in case of a power loss. These exist, and really work. The theory of a pyramid of caching; with "slower and cheaper" at the base and "faster and pricier" towards the top really has been shown to work.
ZFS can do all of this right now, and continuing a little off topic... can also do compressed incremental volume snapshots sent into the cloud :)
Yeah, I do a lot of work with ZFS. All of this stuff really works.
Maybe not science fiction, per se, to some-- but I think it counts. And don't worry, the book is an order of magnitude better than the crappy animated movie which I will not spell out :)
[oops... wrong web site]
'nuff said
None of Amazon, Newegg, Ebay, or even [gasp] Tiger Direct has ever annoyed the @#$%^ out of me for a useless warranty.
Don't make the mistake of confusing "the world I want" with "the world that is". Facebook is a private enterprise, it is free, and it is deeply flawed.
I wish I could park my fancy convertible outside the liquor store with the engine running... and it still be there when I get back. Alas, I must suffer through the world that is. And someone is going to steal my car if I make it too easy.
Who thinks Facebook is private? The whole point is to *not* be private, right? Otherwise... what is the point of Facebook?
If the FBI was going to start monitoring encrypted email, VPNs, and other things where you are *trying* to be private, I would be concerned (yes, I know-- whole 'nuther can o' worms). But Facebook? You are giving the info away as a user, that is the purpose of having a Facebook account.
What does "digital is the future of buying games" mean? Any electronic game I have ever purchased since the 1980's has been digital.
I just got an A77, and I can't understand how anyone could knock the image quality. It has 24 mpixel, I am not impressed with 1:1 blowups detailing every errant pixel.
Having said that, the A77 is probably not what the poster is looking for, and if the NEX7 is half the camera the A77 is (and it is probably more like 90%, which makes me *almost* regret getting the A77), you won't regret it. Good luck finding one until April :)
Sometimes, you don't want to "look" like a photographer. A high quality EVIL camera allows you to be a little more inconspicuous and still get away with great shots.
If you can find one, get the NEX7, and pick a E mount lens that best suits your needs. Then, get an L to E adapter, and you can use 100's of Leicia L mount lenses. Most will only work in manual focus mode, but the peaking display makes focusing easy and accurate.
I just got an Alpha77, which is probably not what you are looking for. I love it, though, and the NEX7 shares a lot of the same technology and features.
When this whole CarrierIQ thing got started, I thought this was simply a diagnostic tool, yet conspiracy theorists were going to jump all over the "...but they could" aspects of the system. I also thought it was a shame, since the carriers and manufacturers ought to be able to monitor the system so they can improve it.
Your post has made me re-think all of my notions. I don't believe any nefarious purpose was afoot-- this was a tool intended to diagnose infrastructure and device performance. However, installing it as a rootkit is a bad call. It provides a vehicle for malware, and a description of its operation-- however technically accurate it migh be-- touches too many evil buzzwords. Such a tool, while useful, is eventually too easy to turn into a PR nightmare (obviously). Then throw the malware hijacking aspect in for good measure.
Verizon does it right, at least as we see it now. Those vans do a great job of real-world testing, where their test equipment is gathering the same metrics as CarrierIQ's software, but with test data nobody will whine about.
But...
At least its not Keanu Reeves.
Screw it... just use perl.
If you want Dtrace and ZFS, just go with FreeBSD. You get pf and jails thrown in for the effort.
http://www.witopia.net/
Pick your country and protocol. Works with the BBC, Hulu, Vimeo, Pandora, Netflix, probably just about everything else. Of course, you have to pick a US proxy for Hulu, a British proxy for the BBC, etc.
Many FOSS projects have their primary development done on xBSD, so that fear may be overstated:
http://uptime.netcraft.com/up/graph?site=www.apache.org
I could not agree more. I have used many of the GEOM features, they all work perfectly. The partitioning scheme, in particular GPART and GLABEL, make working with USB and other hot-swappable devices much easier and less error prone.
The FreeBSD motto is "the power to serve", linux laptop users should keep that in mind.
I had not seen that article yet, thanks. However, it still misses the point. ZFS is *not* intended for 4GB workstations with single hard drives!
I'd like to see them benchmark a ZFS setup with 3 or 4 RAIDz pools concatenated into a 10 to 20 TB partition spanning 10 to 15 physical drives, with an MLC L2ARC. Throw 48GB of ram and an 8 core processor at it as well. Next, properly tune the InnoDB and logging partitions. Then, benchmark MySQL. Finally, try a large update/join spanning many tables, using foreign keys and stored procedures-- and:
-- pull the power plug, see if there is any data corruption
-- do a ZFS snapshot, live, see if the benchmark notices the difference
-- add more RAIDz pools to the live file system
People choose ZFS for the reasons in this article:
http://www.zdnet.com/blog/storage/zfs-data-integrity-tested/811
Sorry to hype on ZFS so much, but that is where my expertise is. Their benchmark was testing a scenario that nobody would be encouraged to ever actually use in the real world. It makes me seriously doubt the usefulness of their other benchmarks (like HURD) when they make such an obvious (to me) mistake as to omit these important details.
Phoronix has a history of questionable choices for their benchmark setups. Hardware, versions, and tuning are... cleverly chosen, almost as if there was a preconceived agenda with inevitable results. Not that there is one-- just like it seems like there is. And so colorfully presented! I remember when they tested ZFS on an i386 version of FreeBSD on a 1G laptop! Others have also noticed this Phoronix phenomenon:
http://forums.freebsd.org/archive/index.php/t-16396.html
http://www.kev009.com/wp/2008/12/phoronix-benchmarking-statistically-significant-and-other-performance-concerns/
The whole point of Hurd, at least right now, is tangential to benchmarks. Nothing wrong with testing, of course, but I think the results should not be used for any long term planning. Nobody is planning on launching a business running on Hurd servers... yet.
How anticlimactic.