Wow! I've always wanted to talk to someone who was running production databases on Fusion IO ioDrives. Can you explain what the storage setup is like? Do you use FusioIO ioDrives in a Tier 1 sort of configuration, backing it with SSDs, or are they totally indepedent? What's the server hardware configuration like? x86? What manufacturer? How many ioDrives per host? What's the total capacity and performance (throughput, IOPS) ? What about the total SSD capacity? Are you using magnetic media for anything in your DB systems? Say, horizontally partitioning out old data and moving to SATA near-line tiers for OLAP, etc? So many question! You should write up a blog post somewhere about your setup, I guarantee there are a lot of people that would be very interested to hear the details!
They claim $20/GB which I have to assume includes more than the bare disks themselves. People routinely pay north of $30/GB for 15K fiber channel storage systems. Lots of things to consider - shelves, hba's, rack enclosures, I/O directors, maybe even power and cooling? Right now Intel SLC drives are over $11/GB and that's just for bare drives. I'd be curious to see if you could build a 50TB RAID5 flash based storage system for under $1M.
Well considering it could easily be hidden in any of their billions of dollars (?) in infrastructure line items in their SEC filings I assume they could "hide" it reasonably well, no?
How exactly do you get 250k IOPS out of 700 disks by the way? If we assume 15k rpm 2.5" disks you _might_ PEAK at about 250 IOPS/sec/disk, and thats being very generous. That's only 175k IOPS, and that's assuming straight reads, if you have a high mix of writes to a RAID volume (which could double your writes in RAID 10) you'd dramatically cut that down. By my math you'd need at LEAST 1,000 FC disks to get 250k IOPS.
I'm just curious, how did you come up with 250k IOPS with 700 disks? Short-stroking? I apologize if I'm missing something obvious, I'm not a storage expert by any stretch of the imagination!
Oh, another thing worth considering - heat and power. I have to assume this many disks comes in at a fraction of the power consumption. Is that worth considering in the $/IOPS as well, if we look beyond the CapEx costs of the drives?
Yes! Absolutely, in many regards. New technology springs up, unstandardized, lack of compatibility, people fighting for market share. Pretty similar to the enterprise storage market actually. Everything is a vertically integrated black box again.
I'm sorry but choosing a platform because you already know a language similar to the one used for that platform seems very shortsighted. There are so many other factors, like what types of applications do you want to develop and in what environment would you find more receptive users (or customers) ?
Learn HTML5/CSS/Javascript. The web (and all the flexibility that HTML5 brings) will become the defacto, platform agnostic tool for mobile devices, just like it has for the web.
Typically for every AMD CPU there is a comparable Intel chip thats either slightly faster and slightly more expensive or slightly slower and slightly less expensive. The one place AMD really is doing well seems to be at the absolute bottom end of the market. If you need a CPU that is literally as inexpensive as you can get, them AMD is worth considering.
Oh you literally meant a beowulf cluster, that's a horrible idea who does that? I'm against taking a workload on bare metal, creating multiple VMs on that same hardware and running identical workloads across all those guests. At least for performance reasons. I've seen people do things like ms exchange clusters in a box (1 server 2 VMs each running exchange enterprise) and there's at least an argument to be made there, but of course it isn't one for performance.
I'm a Cox customer and I get max advertised speed every single day without exception and never have a single problem. Maybe I'm just lucky I don't know but I get well over 20mb/s all the time and we're all DOCSIS 2.0. I'm only supposed to get over 15mb/s with "speedboost" but it works far more often than I _ever_ expected.
Uh, no. Load average can also refer to outstanding I/O (think disk, network, etc). CPU utilization could be 2% but you could have a load average of 500.
There are numerous examples where multiple clustered VMs will perform better than a single OS image, on the same hardware.
While I agree that there are numerous advantages to taking a single host performing many functions and splitting it into many VMs performing individual functions (UNIX philosophy afterall - do one thing and do it well!) but I have yet to find an instance that bears out that statement you made above. What workload would perform better in that instance? If I ran one webserver on bare metal or 4 VMs each running an instance of that same webserver on the same hardware, it will perform better on bare metal (due to hypervisor overhead, if nothing else).
Can you provide a link to a study that proves what you're describing?
I swear, some in the trade rags seem to honestly think there is a benefit to splitting a server into 16 VMs and then combining those into a virtual beowulf cluster for production work (it makes perfect sense for development and testing, of course).
There are a number of benefits to what you just described.
Application compatability issues - much easier to troubleshoot a problem when each guest is performing a single function
Performance - you can move VMs around without rebooting using vMotion if one VM begins to consume too many resources
Upgrades/Maintenance - you can upgrade software or even a single host OS and reboot a guest without affecting the other guests.
In practice these are exceptionally useful features and worth the small performance hit (5-10% hvm overhead) in an enterprise environment.
That's wait jail is for. Normally the company is incorporated and they liquidate it and dissolve the company. Great. What deterrent is that? Unless you can "pierce the corporate veil" and hold the owner of the company personally liable what have you really done? They got to live for a few years making great money. If they live in a homestead state you can't touch that. If they're smart they've hidden a lot more money that the government won't be able to find. Put people in JAIL. That's what it's for.
I agree with you except I feel it's Verizon that put ITSELF in an uncomfortable position. Verizon is pimping out the trust in their name for a price. Simple as that.
Wow, where to start. The point of a trusted CA is you have a 3rd party who can give you the thumbs up or thumbs down on a resource. You seem to work from the assumption that you intrinsically trust the site you're trying to reach. The point of a CA is to verify that you are indeed talking to who you think you're talking to, and to encrypt that communication. If I'm on the Internet and I click a link to xyz-company-ive-never-visited.com and I get an unsigned SSL certificate warning you're saying I should trust that MORE than if I accessed the same site and got a signed SSL certificate from a trusted root CA?
It's based on a chain of trust, not just identity. Just because I can validate Bob the Hacker doesn't mean I should start letting him sign certs and make him a trusted CA. Microsft and Mozilla can't fix this because the trust is delegated by Verizon. So either they remove Verizon's certificates or they can't stop it.
Slashdot Mirror
Wow! I've always wanted to talk to someone who was running production databases on Fusion IO ioDrives. Can you explain what the storage setup is like? Do you use FusioIO ioDrives in a Tier 1 sort of configuration, backing it with SSDs, or are they totally indepedent? What's the server hardware configuration like? x86? What manufacturer? How many ioDrives per host? What's the total capacity and performance (throughput, IOPS) ? What about the total SSD capacity? Are you using magnetic media for anything in your DB systems? Say, horizontally partitioning out old data and moving to SATA near-line tiers for OLAP, etc? So many question! You should write up a blog post somewhere about your setup, I guarantee there are a lot of people that would be very interested to hear the details!
They claim $20/GB which I have to assume includes more than the bare disks themselves. People routinely pay north of $30/GB for 15K fiber channel storage systems. Lots of things to consider - shelves, hba's, rack enclosures, I/O directors, maybe even power and cooling? Right now Intel SLC drives are over $11/GB and that's just for bare drives. I'd be curious to see if you could build a 50TB RAID5 flash based storage system for under $1M.
Well considering it could easily be hidden in any of their billions of dollars (?) in infrastructure line items in their SEC filings I assume they could "hide" it reasonably well, no?
How exactly do you get 250k IOPS out of 700 disks by the way? If we assume 15k rpm 2.5" disks you _might_ PEAK at about 250 IOPS/sec/disk, and thats being very generous. That's only 175k IOPS, and that's assuming straight reads, if you have a high mix of writes to a RAID volume (which could double your writes in RAID 10) you'd dramatically cut that down. By my math you'd need at LEAST 1,000 FC disks to get 250k IOPS.
I'm just curious, how did you come up with 250k IOPS with 700 disks? Short-stroking? I apologize if I'm missing something obvious, I'm not a storage expert by any stretch of the imagination!
Oh, another thing worth considering - heat and power. I have to assume this many disks comes in at a fraction of the power consumption. Is that worth considering in the $/IOPS as well, if we look beyond the CapEx costs of the drives?
I believe the article claimed a 4x performance increase.
Is it the 1980s all over again?
Yes! Absolutely, in many regards. New technology springs up, unstandardized, lack of compatibility, people fighting for market share. Pretty similar to the enterprise storage market actually. Everything is a vertically integrated black box again.
Go ask Palm how licensing their mobile OS worked out for them.
How about we ask Google how it's going for Android instead?
Google up 5 points and Apple _down_ 1.3. If you've been following the trends you'd see that the momentum has been rapidly shifting to Android phones.
I'm sorry but choosing a platform because you already know a language similar to the one used for that platform seems very shortsighted. There are so many other factors, like what types of applications do you want to develop and in what environment would you find more receptive users (or customers) ?
Learn HTML5/CSS/Javascript. The web (and all the flexibility that HTML5 brings) will become the defacto, platform agnostic tool for mobile devices, just like it has for the web.
Typically for every AMD CPU there is a comparable Intel chip thats either slightly faster and slightly more expensive or slightly slower and slightly less expensive. The one place AMD really is doing well seems to be at the absolute bottom end of the market. If you need a CPU that is literally as inexpensive as you can get, them AMD is worth considering.
It's crazy to me that (for people with decent mics at least) Ventrillo sounds better than corporate conference calls.
Woah, maybe better than YOUR corporate conference calls, but definitely not all. Apparently you've never heard of "HD Voice".
Oh you literally meant a beowulf cluster, that's a horrible idea who does that? I'm against taking a workload on bare metal, creating multiple VMs on that same hardware and running identical workloads across all those guests. At least for performance reasons. I've seen people do things like ms exchange clusters in a box (1 server 2 VMs each running exchange enterprise) and there's at least an argument to be made there, but of course it isn't one for performance.
Love it, can't get enough. I'd like to get two or three Cox at once if I could.
I'm a Cox customer and I get max advertised speed every single day without exception and never have a single problem. Maybe I'm just lucky I don't know but I get well over 20mb/s all the time and we're all DOCSIS 2.0. I'm only supposed to get over 15mb/s with "speedboost" but it works far more often than I _ever_ expected.
Uh, no. Load average can also refer to outstanding I/O (think disk, network, etc). CPU utilization could be 2% but you could have a load average of 500.
There are numerous examples where multiple clustered VMs will perform better than a single OS image, on the same hardware.
While I agree that there are numerous advantages to taking a single host performing many functions and splitting it into many VMs performing individual functions (UNIX philosophy afterall - do one thing and do it well!) but I have yet to find an instance that bears out that statement you made above. What workload would perform better in that instance? If I ran one webserver on bare metal or 4 VMs each running an instance of that same webserver on the same hardware, it will perform better on bare metal (due to hypervisor overhead, if nothing else).
Can you provide a link to a study that proves what you're describing?
I swear, some in the trade rags seem to honestly think there is a benefit to splitting a server into 16 VMs and then combining those into a virtual beowulf cluster for production work (it makes perfect sense for development and testing, of course).
There are a number of benefits to what you just described.
In practice these are exceptionally useful features and worth the small performance hit (5-10% hvm overhead) in an enterprise environment.
Because Blizzard has really good lawyers? Welcome to America.
That's wait jail is for. Normally the company is incorporated and they liquidate it and dissolve the company. Great. What deterrent is that? Unless you can "pierce the corporate veil" and hold the owner of the company personally liable what have you really done? They got to live for a few years making great money. If they live in a homestead state you can't touch that. If they're smart they've hidden a lot more money that the government won't be able to find. Put people in JAIL. That's what it's for.
That's "vanilla" warcraft. You can't get to outlands or northrend unless you purchase the expansions.
Wrong. I've got the original retail boxes for both Burning Crusader and Lich King on a bookshelf behind me right now.
I agree with you except I feel it's Verizon that put ITSELF in an uncomfortable position. Verizon is pimping out the trust in their name for a price. Simple as that.
Wow, where to start. The point of a trusted CA is you have a 3rd party who can give you the thumbs up or thumbs down on a resource. You seem to work from the assumption that you intrinsically trust the site you're trying to reach. The point of a CA is to verify that you are indeed talking to who you think you're talking to, and to encrypt that communication. If I'm on the Internet and I click a link to xyz-company-ive-never-visited.com and I get an unsigned SSL certificate warning you're saying I should trust that MORE than if I accessed the same site and got a signed SSL certificate from a trusted root CA?
It's based on a chain of trust, not just identity. Just because I can validate Bob the Hacker doesn't mean I should start letting him sign certs and make him a trusted CA. Microsft and Mozilla can't fix this because the trust is delegated by Verizon. So either they remove Verizon's certificates or they can't stop it.
"He who would trade liberty for some temporary security, deserves neither liberty nor security." - Benjamin Franklin
It starts with tapping your phones. Where does it stop?