Please send in your paypal address, and i will have the $300 promptly invoiced from your account. Reading this message incurs $600 service charge. If you want to expedite this process, please send the money directly to abc@abc.com, which will incur to an extra 200$ self-service charge. Payment is due 14days from today, and will incur an 15% annual interest if not paid in time.
"how brilliant it would be" == "how bloated it would be" "imagine all those ideas" == "imagine all the useless features"
All i want at least is a basic browser which is fast and non-bloated. even chrome is starting to feel a bit bloated..:( and this is on a high end workstation
Yeah, i'm the odd one in the bunch, running so many things at once, but seriously, the cpu cycles, IOPS and RAM is better used elsewhere than "having useless features available", or maintaining bloat
My experience for Ubuntu on desktop was EXACTLY vice-versa, any I/O activity and even sounds started to jitter. I tested different schedulers, read ahead values etc. and that remained, nothing helped. I/O performance was crappy at best for desktop usage.
And who says microsoft can't be good, and do good deeds?:) Yet again MS is doing good deeds, helping those in need. Gates might not be in the helm anymore, but no one can argue that Gates wouldn't be a humanitarian.
I know, this is/. and MS bashing is forte, but yeah, i'm kinda MS fanboy when it comes to desktop, and work productivity:) Nevermind, they actually do make GOOD hardware too, keyboards are way better than Logitech's for example. 2 best keyboards i've ever used have been Microsoft products, and they are overwhelmingly in their own class. I just sold a highend logitech keyboard i had been using for couple weeks, because it sucked compared to low end microsoft keyboard. Some keys in odd positions was the main reasons.
I keep trying linux at desktop, but it fails there utterly time after time again. But i wouldn't put windows as a server, unless AD, Exchange, SharePoint are required. AD, Exchange and SharePoint are excellent pieces of software.
Long story short, i'm very glad MS is doing this, enabling nonprofits to legally enjoy the superior productivity offered for getting their work done, and thus getting protected from using piracy as an excuse to oppress. Sure, one of the points on doing this probably was goodwill PR, but it doesn't negate the fact they are doing this.
I'm now in process to build a VM cluster with 140 drives, and total initial throughput of 50+Gbps. 5Gbps+local cache per node accessible, plenty in our usage! The nodes can be upgraded to have 20+Gbps access to storage, if needed, albeit with "highish" cost. Most of this with commodity hardware to have low costs vs. performance:) 20+k IOPS is the expected real world storage performance, which will be distributed likely to only 50 guest VMs. Which again, is plenty for our usage, when typical VM will be using ~100-250IOPS average, and those demanding VMs with 1k+ requirement will keep running fine:)
The beauty of the designed infrastructure is that everything is trivial to upgrade once the infra is initially built, just add hardware, make couple small configurations (15mins + testing) and done:) Designed it to scale to up to 57 nodes without significant jump in hardware pricing, and even after that, it's only a small restructuring of the infra, without über expensive hardware. We can easily add storage performance in 5-7k IOPS increments, for up to 360k IOPS before infra or drive type needs to be upgraded. Cost of that 360k IOPS would be roughly 240k eur, and we would have multilayer redundant storage capacity of 1.2Pb, and still have easy ways to either upgrade IOPS or capacity, depending upon needs. I guess that would be plenty for the 250-600 VMs we are going to run in the cluster:) (Yes really, likely not more than that). But we are starting small, just that 140 drives.
But how did you build your server? 150 drives in a single server with 9GB/s sounds kind of extreme and hard to design:) Tons of SAS RAID controllers and Multiplier backplanes, on a huge chassis?
16 lane PCIe V3 is 16GB/s, or 128Gbps to compare oranges to oranges. The more common V2, 16 lane is 8GB/s, or single lane 500MB/s=4000Mbps. So a basic x4 PCIe connector would be more than able to handle that speed.
Individual nodes does not matter that much after a certain point. I say shoot for the best combination of power consumption & space, which offers you lowest cost per N amount of performance, and distribute the hell out of it:)
Something is wrong when DB is handling that... The OS underneath should do these conclusions and optimize based on type of storage, without intervention of the DB software.
Of course, applications to have to manage the load they do to a degree, but down to hardware level? That's simply too much, better trust kernel to make the right decisions! Then again, the world isn't perfect...
We have to battle with profoundly bad HDD IO management on the software (Still that software is best for our business), but it is more to do with the complete lack of caching mechanisms of any real tangible value, just the bare minimum. Fortunately, this too is about to be fixed:) We tried an alternative, it was the other extreme, spending as much RAM & CPU as possible to avoid HDD trashing, and the total throughput was multiple times lower:O
SAS can deliver 6Gbps, as can SATA nowadays too (Tho rare). Fastest SSDs hit this limit, but there's no simple way to go beyond (PCIe controller).
SANs have their own bottleneck: SAN switches, which due to "centralized nature" (all traffic from all nodes goes through certain set of switches, or single switch) lowers the overall throughput.
There are ways to have waaaay more IOPS, and waaaay higher throughput total, for way less money. These ways are what we intend to use in our VM cluster to be brought up next spring. We are initially targeting only about 20k IOPS, with throughput scalable upto 20Gbps per node (starting throughput is 5Gbps per node to storage, and upgraded as needed). As this is VMs, throughput is not as important as IOPS. Only on special cases we will be using SSDs. The infra has been designed so that if we maximize a node, the FSB BW will eventually be a potential bottleneck. When we get this cluster online, it's going to be really fun to make some storage tests and see how much IOPS can we achieve and total throughput:D
Efficiently meaning that if middle management or lower management did something bad in the name of the company, the owner would be responsible without giving authorization to said act.
The way it works in US is corrupt & bad. But here in Europe, corporations seems to generally play nicer and held accountable easier.
Actually google's business is advertising made possible by data mining and datacenter operation. Search just happened to be there first as a data mining tool;)
"Surely 640kb is enough ram for anything you could ever want" or something along those lines;)
Jokes aside, magnetic media is a bit too sluggish even for 100Mbps in our use, there is just too much concurrent threads seeking data, and practically a HDD can do only about 20M/s in our usage, rest of the time is spent seeking!
1Gbps is even harder to fill, a 4 HDD server is not enough.
But even higher BW can be utilized in future, think about 4K video, 4096x2160 pixels, and add 3D to that. 3D effectively doubles the required images. It's still far in the future, 2K was released 2005/2006, and 4K is a newish thing for theatres too. But eventually all that will come to home too, maybe in the next 5years. Pace will keep increasing, and BW requirements along with it.
I find the 15Mbps (actual BW i got) at home a bit too little at times, and am hoping to upgrade to 100Mbps, maybe they roll it out here too in Dec. VDSL2+ EFM is arriving as well, offering well over 100Mbps for short distances. IPTV is also becoming norm around here for paid channels.
I would assume there are already some households where people are cursing at low speeds while ADSL2+ 24Mbps. Only need dad watching telly, daughter surfing youtube and facebook, son playing WoW.
Uhm, yes they do send bits. It's the upper layers which send packets, but on hardware layer it's 1 bit at a time, or with the technique summarized 4 bits at a time (time slot).
In simplest term, you cannot somehow make single signal contain thousands of signals.
1 byte consist of 8 bits, and a packet consists of N bytes (upto 65536 total afaik per packet, headers take 29 bytes if i recall correctly). On a single timeslot they can squeeze in now 4 bits or half a byte, which is quite a bit!
psst. don't tell others about this secret, but a byte consist of 8 bits;) and that different colors are being used already (called wavelength frequencys) for oh about atleast 10years?
For 99% of population these add tremendous value. Maybe you prefer the old way of doing things, considered going to library much? Most prefer their content now, immediately. And there's plenty of ways to find it, and immediately without chores. I recommend you learn these basic internet use skills.
Let's pick a simple search for network admins, let's assume you've accustomed to only unmanaged switches so far, or if managed some wierd brand or legacy HW. Now you got some sweet new cisco routers and need to learn howto create VLANs to segment the flow of traffic, search "howto setup cisco vlan" in Youtube results in 81 matches. For CCNA which is Cisco training there was 3 870 videos. On other note, let's assume you are new to coding, and want to compile windows programs under linux, found that too from youtube.
So, the learning curve will raise for those who might be tomorrow's lead network designers and coders.
Before you say "we don't need youtube kiddiez to do the work of professionals!", you might look at current or past generations, how did they get started...
What stuns me is that they are basicly saying that nothing in internet is secure, and everything is hackable.
One way digests, strong cryptography, public key cryptography(SSL) etc etc etc.
Which would mean that US govt has, and these individuals know they have, means to hack any current cryptographic method available, and what is to be available within near term. Which sounds just pure bullshit.
The context of the letter should be known to say that Ronald Rivest was against this, was it because of security issues of certain nature, or potential security issues?
I don't see why digests couldn't work as authorization.
When user votes, for his vote a checksum is created using one-way algo (digest) which is formed from:
Session ID, Voter name, Vote result, a unique key given only to voter and known only by voter and govt, date.
Debian torrents seems to be very poorly seeded as well, along with CentOS (occasionally).
So put up Debian, CentOS, maybe some smaller other distros and something from LegalTorrents:)
We do this systematically, by introducing "testing torrents" on new accounts, basicly a basic set of torrents set to be seeded upon new account creation. We also have a server dedicated just to serve these torrents.
The torrents in our basic set seed extremely well now:)
Review your contract or french law. At least here in Finland you cannot host a SMTP server or connect to SMTP server not maintained by your ISP as antispam policy. Some providers supply connection where there's no blocking of SMTP to outside their network, but you don't have legal right to do so even tho it's not stopped, unless you are an ISP.
Slashdot Mirror
Please send in your paypal address, and i will have the $300 promptly invoiced from your account. Reading this message incurs $600 service charge. If you want to expedite this process, please send the money directly to abc@abc.com, which will incur to an extra 200$ self-service charge. Payment is due 14days from today, and will incur an 15% annual interest if not paid in time.
Please, enjoy your day!
Happy doing business with you,
The MAFIAA
Yeah, they got consulted by the MAFIAA :D
"how brilliant it would be" == "how bloated it would be"
"imagine all those ideas" == "imagine all the useless features"
All i want at least is a basic browser which is fast and non-bloated. even chrome is starting to feel a bit bloated .. :( and this is on a high end workstation
Yeah, i'm the odd one in the bunch, running so many things at once, but seriously, the cpu cycles, IOPS and RAM is better used elsewhere than "having useless features available", or maintaining bloat
My experience for Ubuntu on desktop was EXACTLY vice-versa, any I/O activity and even sounds started to jitter. I tested different schedulers, read ahead values etc. and that remained, nothing helped. I/O performance was crappy at best for desktop usage.
eewww, that is disgusting :O :)
Then again, if it helps saving lives....
nevermind even knowing which, why, how.
Imagine the magic be thin pill: a daily capsule of select bacteria, little to no side effects.
And who says microsoft can't be good, and do good deeds? :)
Yet again MS is doing good deeds, helping those in need. Gates might not be in the helm anymore, but no one can argue that Gates wouldn't be a humanitarian.
I know, this is /. and MS bashing is forte, but yeah, i'm kinda MS fanboy when it comes to desktop, and work productivity :) Nevermind, they actually do make GOOD hardware too, keyboards are way better than Logitech's for example. 2 best keyboards i've ever used have been Microsoft products, and they are overwhelmingly in their own class. I just sold a highend logitech keyboard i had been using for couple weeks, because it sucked compared to low end microsoft keyboard. Some keys in odd positions was the main reasons.
I keep trying linux at desktop, but it fails there utterly time after time again. But i wouldn't put windows as a server, unless AD, Exchange, SharePoint are required. AD, Exchange and SharePoint are excellent pieces of software.
Long story short, i'm very glad MS is doing this, enabling nonprofits to legally enjoy the superior productivity offered for getting their work done, and thus getting protected from using piracy as an excuse to oppress.
Sure, one of the points on doing this probably was goodwill PR, but it doesn't negate the fact they are doing this.
That is quite nice :)
I'm now in process to build a VM cluster with 140 drives, and total initial throughput of 50+Gbps. 5Gbps+local cache per node accessible, plenty in our usage! The nodes can be upgraded to have 20+Gbps access to storage, if needed, albeit with "highish" cost. Most of this with commodity hardware to have low costs vs. performance :) 20+k IOPS is the expected real world storage performance, which will be distributed likely to only 50 guest VMs. Which again, is plenty for our usage, when typical VM will be using ~100-250IOPS average, and those demanding VMs with 1k+ requirement will keep running fine :)
The beauty of the designed infrastructure is that everything is trivial to upgrade once the infra is initially built, just add hardware, make couple small configurations (15mins + testing) and done :) Designed it to scale to up to 57 nodes without significant jump in hardware pricing, and even after that, it's only a small restructuring of the infra, without über expensive hardware. We can easily add storage performance in 5-7k IOPS increments, for up to 360k IOPS before infra or drive type needs to be upgraded. Cost of that 360k IOPS would be roughly 240k eur, and we would have multilayer redundant storage capacity of 1.2Pb, and still have easy ways to either upgrade IOPS or capacity, depending upon needs. I guess that would be plenty for the 250-600 VMs we are going to run in the cluster :) (Yes really, likely not more than that). But we are starting small, just that 140 drives.
But how did you build your server? 150 drives in a single server with 9GB/s sounds kind of extreme and hard to design :) Tons of SAS RAID controllers and Multiplier backplanes, on a huge chassis?
16 lane PCIe V3 is 16GB/s, or 128Gbps to compare oranges to oranges. The more common V2, 16 lane is 8GB/s, or single lane 500MB/s=4000Mbps. So a basic x4 PCIe connector would be more than able to handle that speed.
Individual nodes does not matter that much after a certain point. I say shoot for the best combination of power consumption & space, which offers you lowest cost per N amount of performance, and distribute the hell out of it :)
Something is wrong when DB is handling that ... The OS underneath should do these conclusions and optimize based on type of storage, without intervention of the DB software.
Of course, applications to have to manage the load they do to a degree, but down to hardware level? That's simply too much, better trust kernel to make the right decisions! Then again, the world isn't perfect ...
We have to battle with profoundly bad HDD IO management on the software (Still that software is best for our business), but it is more to do with the complete lack of caching mechanisms of any real tangible value, just the bare minimum. Fortunately, this too is about to be fixed :) We tried an alternative, it was the other extreme, spending as much RAM & CPU as possible to avoid HDD trashing, and the total throughput was multiple times lower :O
SAS can deliver 6Gbps, as can SATA nowadays too (Tho rare). Fastest SSDs hit this limit, but there's no simple way to go beyond (PCIe controller).
SANs have their own bottleneck: SAN switches, which due to "centralized nature" (all traffic from all nodes goes through certain set of switches, or single switch) lowers the overall throughput.
There are ways to have waaaay more IOPS, and waaaay higher throughput total, for way less money. These ways are what we intend to use in our VM cluster to be brought up next spring. We are initially targeting only about 20k IOPS, with throughput scalable upto 20Gbps per node (starting throughput is 5Gbps per node to storage, and upgraded as needed). As this is VMs, throughput is not as important as IOPS. Only on special cases we will be using SSDs. The infra has been designed so that if we maximize a node, the FSB BW will eventually be a potential bottleneck. When we get this cluster online, it's going to be really fun to make some storage tests and see how much IOPS can we achieve and total throughput :D
Efficiently meaning that if middle management or lower management did something bad in the name of the company, the owner would be responsible without giving authorization to said act.
The way it works in US is corrupt & bad. But here in Europe, corporations seems to generally play nicer and held accountable easier.
Actually google's business is advertising made possible by data mining and datacenter operation. Search just happened to be there first as a data mining tool ;)
"Surely 640kb is enough ram for anything you could ever want" or something along those lines ;)
Jokes aside, magnetic media is a bit too sluggish even for 100Mbps in our use, there is just too much concurrent threads seeking data, and practically a HDD can do only about 20M/s in our usage, rest of the time is spent seeking!
1Gbps is even harder to fill, a 4 HDD server is not enough.
But even higher BW can be utilized in future, think about 4K video, 4096x2160 pixels, and add 3D to that. 3D effectively doubles the required images. It's still far in the future, 2K was released 2005/2006, and 4K is a newish thing for theatres too.
But eventually all that will come to home too, maybe in the next 5years. Pace will keep increasing, and BW requirements along with it.
I find the 15Mbps (actual BW i got) at home a bit too little at times, and am hoping to upgrade to 100Mbps, maybe they roll it out here too in Dec. VDSL2+ EFM is arriving as well, offering well over 100Mbps for short distances. IPTV is also becoming norm around here for paid channels.
I would assume there are already some households where people are cursing at low speeds while ADSL2+ 24Mbps. Only need dad watching telly, daughter surfing youtube and facebook, son playing WoW.
Yeah, damn proprietary code! We got to reverse engineer *everything*
Uhm, yes they do send bits. It's the upper layers which send packets, but on hardware layer it's 1 bit at a time, or with the technique summarized 4 bits at a time (time slot).
In simplest term, you cannot somehow make single signal contain thousands of signals.
1 byte consist of 8 bits, and a packet consists of N bytes (upto 65536 total afaik per packet, headers take 29 bytes if i recall correctly). On a single timeslot they can squeeze in now 4 bits or half a byte, which is quite a bit!
psst. don't tell others about this secret, but a byte consist of 8 bits ;) and that different colors are being used already (called wavelength frequencys) for oh about atleast 10years?
wow, 75Gb is so tiny! A busy day i might consume that much :O
Nevermind that our server cluster does that in oh .. i dunno, about every 5minutes, probably less.
For 99% of population these add tremendous value. Maybe you prefer the old way of doing things, considered going to library much? Most prefer their content now, immediately. And there's plenty of ways to find it, and immediately without chores. I recommend you learn these basic internet use skills.
Let's pick a simple search for network admins, let's assume you've accustomed to only unmanaged switches so far, or if managed some wierd brand or legacy HW. Now you got some sweet new cisco routers and need to learn howto create VLANs to segment the flow of traffic, search "howto setup cisco vlan" in Youtube results in 81 matches. For CCNA which is Cisco training there was 3 870 videos. On other note, let's assume you are new to coding, and want to compile windows programs under linux, found that too from youtube.
So, the learning curve will raise for those who might be tomorrow's lead network designers and coders.
Before you say "we don't need youtube kiddiez to do the work of professionals!", you might look at current or past generations, how did they get started ...
hmmph, i thought the 2038 was just 32bit thing, integer overflow, and does not affect 64bit systems?
What stuns me is that they are basicly saying that nothing in internet is secure, and everything is hackable.
One way digests, strong cryptography, public key cryptography(SSL) etc etc etc.
Which would mean that US govt has, and these individuals know they have, means to hack any current cryptographic method available, and what is to be available within near term. Which sounds just pure bullshit.
The context of the letter should be known to say that Ronald Rivest was against this, was it because of security issues of certain nature, or potential security issues?
I don't see why digests couldn't work as authorization.
When user votes, for his vote a checksum is created using one-way algo (digest) which is formed from:
Session ID, Voter name, Vote result, a unique key given only to voter and known only by voter and govt, date.
Now crack that one ;)
LegalTorrents.com is your source :)
Debian torrents seems to be very poorly seeded as well, along with CentOS (occasionally).
So put up Debian, CentOS, maybe some smaller other distros and something from LegalTorrents :)
We do this systematically, by introducing "testing torrents" on new accounts, basicly a basic set of torrents set to be seeded upon new account creation. We also have a server dedicated just to serve these torrents.
The torrents in our basic set seed extremely well now :)
Review your contract or french law. At least here in Finland you cannot host a SMTP server or connect to SMTP server not maintained by your ISP as antispam policy. Some providers supply connection where there's no blocking of SMTP to outside their network, but you don't have legal right to do so even tho it's not stopped, unless you are an ISP.