CNA - Converged Network Agora
With every generation of Ethernet there is always a new crop of agile startups who race to silicon and deliver a variety of new ASICs for network adapters and switches. As the market matures competition thins the herd and after several years only the best products remain. This Darwinian process is what the market does best, and the process was in full swing in January of 2009 when I authored a post "Thinning the 10GbE Herd." In this article I mentioned NetXen, Neterion, Tehuti, ServerEngines, NetEffect, & Teak Technologies all of which are gone three years later.
Today three startups who have remained focused on high performance 10GbE network interface card (NIC) silicon & software for the past six years still remain. These companies are: Chelsio, Myricom & Solarflare. Don't get me wrong, large public companies like Intel, Broadcom, Emulex & Qlogic ship far more 10GbE ports, but these are all well established public companies focused on the larger 10GbE market, and with significant silicon production capabilities. To be complete there is also a cache of very niche players like: Napatech, Endace and HotLava but these firms leverage FPGAs or in HotLava's case Intel silicon.
As someone who's been selling 10GbE for the past seven years full time, it finally appears that 2012 will be the year of convergence. Chelsio & Solarflare are both venture capital backed and by most estimates are reaching the end of their funding. Myricom is privately held, and has been running on revenue for 18 years, also last week they announced a partnership with Emulex.
In the recent past Intel gobbled up NetEffect & Folcrum so their appetite for 10GbE appears pretty well satiated. Early on Broadcom acquired Siliquent, and they appear content. So the game of musical chairs has started. Emulex & Qlogic have made various plays, but they both still appear hungry, which leaves only two seats. We have three remaining players (Chelsio, Myricom & Solarflare). Later this year the music will stop, none may remain, and the convergence will be complete. Two may win, and likely one will lose, the question is which?
Then in 2013 we get to start it all over again with 40GbE!
Today three startups who have remained focused on high performance 10GbE network interface card (NIC) silicon & software for the past six years still remain. These companies are: Chelsio, Myricom & Solarflare. Don't get me wrong, large public companies like Intel, Broadcom, Emulex & Qlogic ship far more 10GbE ports, but these are all well established public companies focused on the larger 10GbE market, and with significant silicon production capabilities. To be complete there is also a cache of very niche players like: Napatech, Endace and HotLava but these firms leverage FPGAs or in HotLava's case Intel silicon.
As someone who's been selling 10GbE for the past seven years full time, it finally appears that 2012 will be the year of convergence. Chelsio & Solarflare are both venture capital backed and by most estimates are reaching the end of their funding. Myricom is privately held, and has been running on revenue for 18 years, also last week they announced a partnership with Emulex.
In the recent past Intel gobbled up NetEffect & Folcrum so their appetite for 10GbE appears pretty well satiated. Early on Broadcom acquired Siliquent, and they appear content. So the game of musical chairs has started. Emulex & Qlogic have made various plays, but they both still appear hungry, which leaves only two seats. We have three remaining players (Chelsio, Myricom & Solarflare). Later this year the music will stop, none may remain, and the convergence will be complete. Two may win, and likely one will lose, the question is which?
Then in 2013 we get to start it all over again with 40GbE!
FPGAs on 10GbE NICs, An Idea Who's Time Has Passed
20/04/12 15:11 Filed in: HFT
A few months ago SolarFlare announced a new class of Network Interface Card (NIC), a hybrid adapter, that will be available in June. This hybrid combines their generic 10GbE ASIC with a Field Programmable Gate Array (FPGA) chip, some RAM and then they wrap all this hardware in a Software Development Kit (SDK). This will then be marketed as a complete solution for the High Frequency Trading (HFT) market. Rumors exist that they'll also try and sell it into the network security market, and perhaps others.
At the time of this writing high performance dual port NICs have a street price between $550 & $750, this new hybrid NIC is rumored to cost ten times this. So why would someone even consider this approach? Simple to reduce load on the host CPU cores. The initial pitch is that this hybrid will take on the role of the feed handler. Typically a feed handler runs on several cores of a multi-core server today. It receives trading data from all the relevant exchanges, then filters off all the unwanted information, normalizes what remains and then passes this onto cores running the algorithmic trading code. By freeing up the feed handler cores, through the use of a hybrid NIC, this processing power can be allocated to run more advanced algorithmic codes.
On the surface the pitch sounds like a great idea. Use a proven low-latency ASIC to pull packets off the wire, send the boring packets on to the OS and the interesting financial stuff to the FPGA. It's when you get into the details that you realize it's nothing more than a marketing scheme. When this product was designed I'm sure it sounded like a good idea, most 1U and 2U servers had eight cores and systems were getting CPU bound. As this NIC hits the market though Intel has once again turned the crank and vendors like IBM and HP are now delivering dual socket 16 core, 32 thread servers that will easily pickup the slack. A nicely configured HP DL360P with 16 cores, 32GB memory, etc… is available today for $10K, adding one of these hybrid NICs will nearly double the hardware price of your trading platform. Note, this before you even crack open the SDK and hire the small army of consultants you'll need to program the FPGA.
Typically we've found that the normal packet flow from multiple exchanges into a trading server is roughly 200-300K packets per second (pps), with very rare bursts up to 800K. So if one were to set aside four cores for feed handling, with an average feed load of 250Kpps, and assuming the feeds were evenly distributed each core would have 16 microseconds per packet. On these new 2.2Ghz Intel E5 systems this translates to roughly 8K instructions per packet to filter & normalize. This assumes two threads per core and an average of four clock ticks per instruction.
Like TCP Offload Engines (TOEs) these hybrid NICs sound great when they're first proposed, but on in-depth analysis and particularly after Moore's law kicks in, they soon become solutions looking for a problem, a novelty. With Intel's new E5s why would anyone seriously invest their time hardware & consulting budgets on an outdated approach?
At the time of this writing high performance dual port NICs have a street price between $550 & $750, this new hybrid NIC is rumored to cost ten times this. So why would someone even consider this approach? Simple to reduce load on the host CPU cores. The initial pitch is that this hybrid will take on the role of the feed handler. Typically a feed handler runs on several cores of a multi-core server today. It receives trading data from all the relevant exchanges, then filters off all the unwanted information, normalizes what remains and then passes this onto cores running the algorithmic trading code. By freeing up the feed handler cores, through the use of a hybrid NIC, this processing power can be allocated to run more advanced algorithmic codes.
On the surface the pitch sounds like a great idea. Use a proven low-latency ASIC to pull packets off the wire, send the boring packets on to the OS and the interesting financial stuff to the FPGA. It's when you get into the details that you realize it's nothing more than a marketing scheme. When this product was designed I'm sure it sounded like a good idea, most 1U and 2U servers had eight cores and systems were getting CPU bound. As this NIC hits the market though Intel has once again turned the crank and vendors like IBM and HP are now delivering dual socket 16 core, 32 thread servers that will easily pickup the slack. A nicely configured HP DL360P with 16 cores, 32GB memory, etc… is available today for $10K, adding one of these hybrid NICs will nearly double the hardware price of your trading platform. Note, this before you even crack open the SDK and hire the small army of consultants you'll need to program the FPGA.
Typically we've found that the normal packet flow from multiple exchanges into a trading server is roughly 200-300K packets per second (pps), with very rare bursts up to 800K. So if one were to set aside four cores for feed handling, with an average feed load of 250Kpps, and assuming the feeds were evenly distributed each core would have 16 microseconds per packet. On these new 2.2Ghz Intel E5 systems this translates to roughly 8K instructions per packet to filter & normalize. This assumes two threads per core and an average of four clock ticks per instruction.
Like TCP Offload Engines (TOEs) these hybrid NICs sound great when they're first proposed, but on in-depth analysis and particularly after Moore's law kicks in, they soon become solutions looking for a problem, a novelty. With Intel's new E5s why would anyone seriously invest their time hardware & consulting budgets on an outdated approach?
4.0 Microseconds for a Half Round Trip
31/01/12 15:50 Filed in: HFT
From a press release published yesterday:
Myricom Adds Design Win for High Frequency Trading with Wall Street Software and Infrastructure Specialist, Trading Physics
ARCADIA, Calif., Jan 30, 2012 (BUSINESS WIRE) -- --Achieves 4.0 microsecond latency normalizing US equity depth feeds
Myricom, the technology leader in extreme-performance 10-Gigabit Ethernet solutions specialized for vertical markets, today announces that Trading Physics, a producer of analytic software and custom market data delivery systems, has adopted Myricom's 10G-PCIE2-8C2-2S dual port SFP+ adapters along with Myricom's DBL 2.0(TM) low-latency software for High Frequency Trading (HFT) applications.
"The depth feeds contain all orders placed for equities listed on five exchanges: NYSE, NYSE/ARCA, NASDAQ, BATS, and Direct Edge -- an avalanche of data, and processing this order book data with the lowest possible latency is critical to the success of high frequency trading strategies. A customer asked us to achieve sub-five microsecond latency from the top of the switch to client server delivery. After extensive research we found Myricom's extreme performance low-latency solution to be very attractive. Unlike products from other vendors, Myricom's solution did not require the use of custom server hardware to process the feeds," said Artem Novikov, Founder of Trading Physics. "We ordered network adapters, set up a benchmarking test bed using Myricom native DBL libraries, and rapidly surpassed the target benchmark with production data. The customer is delighted with the results."...
Myricom Adds Design Win for High Frequency Trading with Wall Street Software and Infrastructure Specialist, Trading Physics
ARCADIA, Calif., Jan 30, 2012 (BUSINESS WIRE) -- --Achieves 4.0 microsecond latency normalizing US equity depth feeds
Myricom, the technology leader in extreme-performance 10-Gigabit Ethernet solutions specialized for vertical markets, today announces that Trading Physics, a producer of analytic software and custom market data delivery systems, has adopted Myricom's 10G-PCIE2-8C2-2S dual port SFP+ adapters along with Myricom's DBL 2.0(TM) low-latency software for High Frequency Trading (HFT) applications.
"The depth feeds contain all orders placed for equities listed on five exchanges: NYSE, NYSE/ARCA, NASDAQ, BATS, and Direct Edge -- an avalanche of data, and processing this order book data with the lowest possible latency is critical to the success of high frequency trading strategies. A customer asked us to achieve sub-five microsecond latency from the top of the switch to client server delivery. After extensive research we found Myricom's extreme performance low-latency solution to be very attractive. Unlike products from other vendors, Myricom's solution did not require the use of custom server hardware to process the feeds," said Artem Novikov, Founder of Trading Physics. "We ordered network adapters, set up a benchmarking test bed using Myricom native DBL libraries, and rapidly surpassed the target benchmark with production data. The customer is delighted with the results."...
Blogs & eZines
26/01/12 08:59 Filed in: Site
Everything that is interesting is back. Enjoy! More new content is on the way.
Sub-$1K Timecode-based SYNC NIC General Availability
17/01/12 10:38 Filed in: Time
This month our new 10G-PCIE2-8C2-2S-SYNC dual-SFP+-port adapter has become generally available. It is specifically designed for applications such as cybersecurity and high frequency trading where accurate timestamping, lossless packet sniffing and injection at line rate, and cutting edge latencies are required.
“With Myricom’s new SYNC network adapters, we can timestamp all incoming packets with 1us precision, allowing us to more accurately measure and monitor the performance of our high frequency trading platform,” said Oleg Chernyakhovskiy, CTO, Lynx Capital Partners, a leading provider of high volume, low latency automated trading applications.
“Myricom’s new SYNC adapter makes precise timestamping accessible for the broad 10-Gigabit Ethernet market,” said Dr. Nan Boden, CEO of Myricom. “Our solution, by using a CDMA or GPS timecode source, gives our customers the capability to accurately timestamp packets without having to sacrifice 10GbE throughput as is the case with competing solutions.”
*Note most of this text came from our press release.
“With Myricom’s new SYNC network adapters, we can timestamp all incoming packets with 1us precision, allowing us to more accurately measure and monitor the performance of our high frequency trading platform,” said Oleg Chernyakhovskiy, CTO, Lynx Capital Partners, a leading provider of high volume, low latency automated trading applications.
“Myricom’s new SYNC adapter makes precise timestamping accessible for the broad 10-Gigabit Ethernet market,” said Dr. Nan Boden, CEO of Myricom. “Our solution, by using a CDMA or GPS timecode source, gives our customers the capability to accurately timestamp packets without having to sacrifice 10GbE throughput as is the case with competing solutions.”
*Note most of this text came from our press release.