Saturday, February 27, 2021
The next replacement for websockets, UDP.
If you're a dev, please like this post, and "star" the issue below, it's pretty far reaching for the internet.
https://bugs.chromium.org/p/chromium/issues/detail?id=1069742
Tuesday, December 22, 2020
Is oxidative stress, during sleep, killing you?
Can't sleep? Suffer from insomnia? Get agitated when lying in bed?
Ever heard of Pavlov and how he trained his dogs to salivate at the sound of a bell? He managed to train the dogs involuntary response, where they associated food with the ring of a bell, and the unconscious mind would trigger salivation.
Ponder another scenario. If every night, during sleep, someone grabbed your neck and started restricting airflow. A fight/flight response would be created, right? Question, if this happened every night, would the unconcious mind associate your bed with the fight/flight response. And you, the sleeper, have no idea this involuntary association is being created in the unconscious mind.
Oxidative stress, UARS (upper airway restriction), etc. are very real phenomenon. Unconscious conditioning/training is also very real.
Detecting and measuring snoring issues is pretty easy, $10 will get a copy of snorelab. Snorelab measurements are coming in high? Check out Alaxolita nasal stents. They take a month or two to get used too, pretty uncomfortable at first, but your snorelab score will collapse into single digits. From my experience, Alaxolito stents will change your life, in a big way, for the better.
UARS is harder to detect/measure, see this blog for the details.
To learn more about unconscious associations/training we unwittingly do, buy a Dreem2 EEG headset and take the advanced course.
Happy sleeping!
Ever heard of Pavlov and how he trained his dogs to salivate at the sound of a bell? He managed to train the dogs involuntary response, where they associated food with the ring of a bell, and the unconscious mind would trigger salivation.
Ponder another scenario. If every night, during sleep, someone grabbed your neck and started restricting airflow. A fight/flight response would be created, right? Question, if this happened every night, would the unconcious mind associate your bed with the fight/flight response. And you, the sleeper, have no idea this involuntary association is being created in the unconscious mind.
Oxidative stress, UARS (upper airway restriction), etc. are very real phenomenon. Unconscious conditioning/training is also very real.
Detecting and measuring snoring issues is pretty easy, $10 will get a copy of snorelab. Snorelab measurements are coming in high? Check out Alaxolita nasal stents. They take a month or two to get used too, pretty uncomfortable at first, but your snorelab score will collapse into single digits. From my experience, Alaxolito stents will change your life, in a big way, for the better.
UARS is harder to detect/measure, see this blog for the details.
To learn more about unconscious associations/training we unwittingly do, buy a Dreem2 EEG headset and take the advanced course.
Happy sleeping!
Sunday, December 20, 2020
Software development, are we doing something wrong? Check out the New York Stock Exchange, NYSE Pillar.
At risk of oversimplifying, something worth noting. Since the beginning of man, from an engineering perspective, we have made the biggest jumps when building tools which are used to build other tools. Why shouldn't the same thing apply to software. Why not write systems which generate other systems, etc. etc.?
NYSE Pillar, arguably one of the most critical systems on the planet, was built this way. The brains behind it? Alexei Lebedev. I had the privilege of observing this feat while working at Intercontinental Exchange, when it purchased the New York Stock Exchange.
Interested in seeing the goodies? Think we should have an open source framework? Merry Christmas, see OpenACR
NYSE Pillar, arguably one of the most critical systems on the planet, was built this way. The brains behind it? Alexei Lebedev. I had the privilege of observing this feat while working at Intercontinental Exchange, when it purchased the New York Stock Exchange.
Interested in seeing the goodies? Think we should have an open source framework? Merry Christmas, see OpenACR
Friday, December 11, 2020
Sleep study technology is getting better and cheaper
Oxidative stress during sleep, some make a compelling case, bad breathing during sleep is what causes a good number of diseases. If we can detect it cheaply, hopefully lives can be saved. The folks at Dreem2 are doing some incredible work.
Beta breath tracking.
Another option is snorelab
Beta breath tracking.
Another option is snorelab
AF_XDP and the linux kernel, this is gonna be big
I've been talking about kernel bypass for a long time. Now, it's being distributed with the linux kernel. It's now possible to intercept packets before they hit the kernel. Xilinx is now taking advantage of the feature. Amazing times!
See the github repo here.
See the github repo here.
Covid happens
Coronavirus, I thought everyone was overreacting, then it happened to my dad. Think of a panicked person yelling/drowning, trying to catch a breath. That's what it looked like on zoom, for days, before finally passing. He was 76.
I can't help but think, these are very early days, and we still have a lot to learn/discover. Our FDA is the gold standard, with respect to quality and testing, second to none on the planet. And with that being said, the FDA says "oops" via recalls thousands of times per year.
I can't help but think, these are very early days, and we still have a lot to learn/discover. Our FDA is the gold standard, with respect to quality and testing, second to none on the planet. And with that being said, the FDA says "oops" via recalls thousands of times per year.
Sunday, June 10, 2018
The coming era of Neurofeedback
Decided to start meditating? Bought a Muse? Feel like throwing it out the window? Gonna put it on Ebay?
You're not alone, and I'd like to share my experience learning to meditate with Muse, the funny little headband one can purchase on ChooseMuse
For those who don't want to read a lengthy description of my journey, and just want to learn the "trick", here's the scoop. For the first couple weeks of using your Muse, DO NOT close your eyes during the calibration phase. Use the "Rain Forest" soundscape. And calibrate with eyes open until being able to maintain the "bird chirping" for at least thirty minutes per session, over five consecutive days, showing 80% calm or better. Then, start half closing your eyes during calibration, and repeat the above cycle. Eventually there will be enough "mental muscle" to calibrate with your eyes closed, and still find "the birds".
Now I'm sure your saying, what good is the "percent calm" reading if the calibration from which it is derived can change? For those familiar with weight lifting, the best way to view your Muse is as a fantastic barbell. The calibration phase is where the user places weight on the barbell. Keeping ones eyes open during calibration is the equivalent of placing minimal weight on the barbell, so squeezing out eight reps is easy (ie hitting 80%). Closing your eyes during calibration is the equivalent of placing 200 lbs on the barbell, so hitting 80% will be much harder. To throw 1000 lbs on the barbell, an advanced meditation session will involve doing a "warm-up" meditation to enter an increased "calm" state for calibration, end the warm-up session, and then start a new session to enable re-calibration at an increased difficulty level, ie placing more weight on the barbell.
Luckily, I'm part mongoose, might be why my wife claims I've got Aspergers. Once I get my teeth in something, letting go simply isn't an option. The cobra must die! (see Riki Tiki Tavi) Needless to say, I've stuck it out with my Muse.
My first couple months trying to get somewhere with my Muse. Think about trying to "push a cat into the toilet" or trying to "nail Jello to a tree". Both being extremely frustrating activities. But after substantial repeated effort I started "sensing" something going on in my head. Sometimes it would start on the right side, move around a bit, and then go to the left side. As time progressed, going into three to four months, the feeling became much more focused resulting in a very "tight" sensation between both my eyes and up a bit. Some might mistake it for a light headache, but it's not. After getting "Muse Direct" to observe all my brain waves, I've come to realize the presence of prodigious "alpha waves" at the front, right/left sides, and back of my head. This is a state typically only achieved by "advanced" meditation gurus after decades of preparation/training.
Neural bridging. In the same way muscles get larger when lifting weights, neural bridging is the strengthening of the "calm" muscle in your brain. See The Automatic Self on Amazon. Continuing with Muse training, I am sensing an increased ability to enter into a "calm" state with minimal effort. Soon I suspect it'll be easy as "falling off a log".
One other tip, another discovery with Muse has been how important "breath" can be to entering a calm state. Research "belly breathing" on google, this is very key to focused meditation. And will accelerate your journey with Muse.
At this point, all I can say is Namaste! Hang in there, it's worth making the commitment to visit the gym and exercise your brain every day. I actually do it twice a day if I can pull it off. Currently, my Muse points stand at 1,462,616. After all the frustration, I'm glad I stuck with it. This quirky little headband is changing my life experience. In the same way physical training, can be uncomfortable, meditation can be the same. "Lean" into it. In the way muscles get stronger, your brain does the same.
Neurofeedback/biofeedback is for real. I've been humming under trees, etc. for a couple decades now, only to find myself holding an empty bag. With Muse I now know 100%, I'm meditating. The experience is clear as "getting hit in the head with a brick", no doubt whatsoever.
Remember learning to ride a bike? Incredibly frustrating at first, but once that first ten to twenty feet is traversed without your feet touching the ground, the notion is born, "I can do this thing!". My journey with Muse has turned me into a curious playful kid again. And just like I could learn to ride a bike, thanks to Muse, I can meditate as well!
Feeling a bit sad lately, in the realization many of today's health issues are "rooted" in one having anxiety, fear, etc. Mind, composed of the nervous system and brain is what "drives" everything in the body. Many people are needlessly suffering. Pretty soon, I fully expect Neurofeedback will be getting prescribed as a treatment for anything ranging from insomnia to erectile dysfunction and obesity. No more need for Lunesta, Viagra, Cialis, etc. A golden age for humanity is just beginning.
I'll try and report back again in a few months. Stay strong and "hum" my friends!
You're not alone, and I'd like to share my experience learning to meditate with Muse, the funny little headband one can purchase on ChooseMuse
For those who don't want to read a lengthy description of my journey, and just want to learn the "trick", here's the scoop. For the first couple weeks of using your Muse, DO NOT close your eyes during the calibration phase. Use the "Rain Forest" soundscape. And calibrate with eyes open until being able to maintain the "bird chirping" for at least thirty minutes per session, over five consecutive days, showing 80% calm or better. Then, start half closing your eyes during calibration, and repeat the above cycle. Eventually there will be enough "mental muscle" to calibrate with your eyes closed, and still find "the birds".
Now I'm sure your saying, what good is the "percent calm" reading if the calibration from which it is derived can change? For those familiar with weight lifting, the best way to view your Muse is as a fantastic barbell. The calibration phase is where the user places weight on the barbell. Keeping ones eyes open during calibration is the equivalent of placing minimal weight on the barbell, so squeezing out eight reps is easy (ie hitting 80%). Closing your eyes during calibration is the equivalent of placing 200 lbs on the barbell, so hitting 80% will be much harder. To throw 1000 lbs on the barbell, an advanced meditation session will involve doing a "warm-up" meditation to enter an increased "calm" state for calibration, end the warm-up session, and then start a new session to enable re-calibration at an increased difficulty level, ie placing more weight on the barbell.
Luckily, I'm part mongoose, might be why my wife claims I've got Aspergers. Once I get my teeth in something, letting go simply isn't an option. The cobra must die! (see Riki Tiki Tavi) Needless to say, I've stuck it out with my Muse.
My first couple months trying to get somewhere with my Muse. Think about trying to "push a cat into the toilet" or trying to "nail Jello to a tree". Both being extremely frustrating activities. But after substantial repeated effort I started "sensing" something going on in my head. Sometimes it would start on the right side, move around a bit, and then go to the left side. As time progressed, going into three to four months, the feeling became much more focused resulting in a very "tight" sensation between both my eyes and up a bit. Some might mistake it for a light headache, but it's not. After getting "Muse Direct" to observe all my brain waves, I've come to realize the presence of prodigious "alpha waves" at the front, right/left sides, and back of my head. This is a state typically only achieved by "advanced" meditation gurus after decades of preparation/training.
Neural bridging. In the same way muscles get larger when lifting weights, neural bridging is the strengthening of the "calm" muscle in your brain. See The Automatic Self on Amazon. Continuing with Muse training, I am sensing an increased ability to enter into a "calm" state with minimal effort. Soon I suspect it'll be easy as "falling off a log".
One other tip, another discovery with Muse has been how important "breath" can be to entering a calm state. Research "belly breathing" on google, this is very key to focused meditation. And will accelerate your journey with Muse.
At this point, all I can say is Namaste! Hang in there, it's worth making the commitment to visit the gym and exercise your brain every day. I actually do it twice a day if I can pull it off. Currently, my Muse points stand at 1,462,616. After all the frustration, I'm glad I stuck with it. This quirky little headband is changing my life experience. In the same way physical training, can be uncomfortable, meditation can be the same. "Lean" into it. In the way muscles get stronger, your brain does the same.
Neurofeedback/biofeedback is for real. I've been humming under trees, etc. for a couple decades now, only to find myself holding an empty bag. With Muse I now know 100%, I'm meditating. The experience is clear as "getting hit in the head with a brick", no doubt whatsoever.
Remember learning to ride a bike? Incredibly frustrating at first, but once that first ten to twenty feet is traversed without your feet touching the ground, the notion is born, "I can do this thing!". My journey with Muse has turned me into a curious playful kid again. And just like I could learn to ride a bike, thanks to Muse, I can meditate as well!
Feeling a bit sad lately, in the realization many of today's health issues are "rooted" in one having anxiety, fear, etc. Mind, composed of the nervous system and brain is what "drives" everything in the body. Many people are needlessly suffering. Pretty soon, I fully expect Neurofeedback will be getting prescribed as a treatment for anything ranging from insomnia to erectile dysfunction and obesity. No more need for Lunesta, Viagra, Cialis, etc. A golden age for humanity is just beginning.
I'll try and report back again in a few months. Stay strong and "hum" my friends!
Saturday, May 26, 2018
Is hell a place? Or a state of mind?
Figure it out.
Thinking, Fast and Slow
Neurofeedback and taming the Amygdala
Understanding your Amygdala
The Amygdala and how it relates to next generation management
Currently in the Abyss? Start here!
Believe it or not, these all "tie in" together. Solve the riddle!
Thinking, Fast and Slow
Neurofeedback and taming the Amygdala
Understanding your Amygdala
The Amygdala and how it relates to next generation management
Currently in the Abyss? Start here!
Believe it or not, these all "tie in" together. Solve the riddle!
Sunday, January 28, 2018
Howto fix the Spectre and Meltdown exploits
In prod, use bare metal servers which you completely control. Anytime there is a chance of someone running a VM on the same physical server as yourself, which is out of your control, the threat surface exists. Keep it simple, and ensure you have complete control of the bare metal server.
AWS is in the process of providing bare metal options. And www.packet.net has already been doing this for quite some time.
Friday, January 5, 2018
Performance workaround for Meltdown and Spectre exploits/vulerabilities
Those of us familiar with kernel-bypass techniques are keenly aware of the overhead in the user/kernel transition, especially when multiplexing many thousands of tcp sockets.
Believe it or not, Windows already has the pieces needed to recover the lost performance, due to the new security fix overhead found in the user/kernel transition when doing socket i/o. We can create what is essentially a user-mode tcpip stack using Windows RIO (aka Registered i/o) to batch many tcpip packets into a single user/kernel transition. Combine with the User Mode Scheduler (UMS), which basically intercepts i/o calls, before incurring the overhead of the user/kernel transition, and batches these calls into our RIO based tcpip engine.
This infrastructure could then be a shim layer within windows docker containers running both windows and linux.
What we need now is a UMS concept on the linux side so the same can be done.
BTW, Intel CEO Brian Krzanich needs to go asap. Given the stock sale, he is doing more harm than good.
Believe it or not, Windows already has the pieces needed to recover the lost performance, due to the new security fix overhead found in the user/kernel transition when doing socket i/o. We can create what is essentially a user-mode tcpip stack using Windows RIO (aka Registered i/o) to batch many tcpip packets into a single user/kernel transition. Combine with the User Mode Scheduler (UMS), which basically intercepts i/o calls, before incurring the overhead of the user/kernel transition, and batches these calls into our RIO based tcpip engine.
This infrastructure could then be a shim layer within windows docker containers running both windows and linux.
What we need now is a UMS concept on the linux side so the same can be done.
BTW, Intel CEO Brian Krzanich needs to go asap. Given the stock sale, he is doing more harm than good.
Friday, April 14, 2017
Diagnose sleep apnea for cheap
Over a year ago I did the take home sleep study, cost around $300. It came back negative for sleep apnea. Turns out the study was wrong.
The more reliable, high-end sleep studies which can detect sleep apnea and UARS (upper airway restriction syndrome), cost thousands of dollars. Most CPAP machines these days track AHI (sleep apnea index) and cost roughly $300. So, I asked my doctor for a CPAP prescription, paying cash for the machine. Best choice I ever made. After using CPAP for a couple months the measurements showed I have mild to moderate sleep apnea. And I'm now getting twice as much sleep.
I'm 6-2, weighing 190 lbs. Not an overweight person. But, I have a narrow neck and crowded teeth. Both markers for UARS. UARS is very expensive/challenging to diagnose. So, if you're tired all the time and suspect the doc is missing something, might be time to get with your ENT and just buy a cpap machine. Mine is a REMStar Auto (System One 60 Series). Best purchase I ever made. Granted, acclimating to CPAP is no small task, took me a couple months, but well worth the struggle. Also finding a mask can take time, I settled on the Simplus mask.
Another amazing thing has happened in the last few weeks. The high-end sleep study I mentioned earlier is now done every night in your fitbit with the latest software update. All sleep stages can now be tracked. Yet another great way to track/measure if you have UARS or sleep apnea. Incredible times we live in, such a great time to be alive!
Many thanks to Doctor Steven Park, his blogging on UARS and sleep apnea have changed my life.
The more reliable, high-end sleep studies which can detect sleep apnea and UARS (upper airway restriction syndrome), cost thousands of dollars. Most CPAP machines these days track AHI (sleep apnea index) and cost roughly $300. So, I asked my doctor for a CPAP prescription, paying cash for the machine. Best choice I ever made. After using CPAP for a couple months the measurements showed I have mild to moderate sleep apnea. And I'm now getting twice as much sleep.
I'm 6-2, weighing 190 lbs. Not an overweight person. But, I have a narrow neck and crowded teeth. Both markers for UARS. UARS is very expensive/challenging to diagnose. So, if you're tired all the time and suspect the doc is missing something, might be time to get with your ENT and just buy a cpap machine. Mine is a REMStar Auto (System One 60 Series). Best purchase I ever made. Granted, acclimating to CPAP is no small task, took me a couple months, but well worth the struggle. Also finding a mask can take time, I settled on the Simplus mask.
Another amazing thing has happened in the last few weeks. The high-end sleep study I mentioned earlier is now done every night in your fitbit with the latest software update. All sleep stages can now be tracked. Yet another great way to track/measure if you have UARS or sleep apnea. Incredible times we live in, such a great time to be alive!
Many thanks to Doctor Steven Park, his blogging on UARS and sleep apnea have changed my life.
Monday, April 10, 2017
Excellent H1B article from the IEEE
What the U.S. desperately needs right now is inflation. Inflation cannot get a foothold until wages start going up. Reigning in the H1B program and cutting down on abuse fostered by large corporations is the key to getting wage inflation. Otherwise, we need to start talking about UBI, Universal Basic Income.
See IEEE article here.
Saturday, April 8, 2017
Federal H1B abuse tip line
Tuesday, April 4, 2017
Low latency, high throughput TCP
For C++ devs looking to push the envelope with respect to latency and throughput when using TCP, checkout mTCP!
https://github.com/eunyoung14/mtcp
Tech wages continue to be depressed thanks to Steve Jobs wage fixing pact
The damages awarded thus far come nowhere close to addressing the damage done to the tech industry overall. Add in H1B's to enforce cheap labor, and the future still looks dim for techies. Will Trump keep his promise?
https://www.theregister.co.uk/2017/02/02/disney_pixar_lucasfilm_wage_fixing/
http://www.breitbart.com/big-government/2016/11/28/obama-expands-h1b-program/
Tuesday, March 14, 2017
Async and await on the JVM
For those interested, this is a very very big deal. Many of us from the C# side of things have been longing for async await on the JVM. While I realize Quasar has been around for a while, Kotlin has a much more elegant compiler supported approach. This is awesome!!
https://blog.jetbrains.com/kotlin/2017/03/kotlin-1-1/
https://github.com/Kotlin/kotlin-coroutines
https://github.com/Kotlin/kotlinx.coroutines
Given Kotlin will support native code compilation in the future on everything from IOS, Android to the PC, the possibilities for having a single build which targets all platforms are getting closer.
https://blog.jetbrains.com/kotlin/2017/03/kotlin-1-1/
https://github.com/Kotlin/kotlin-coroutines
https://github.com/Kotlin/kotlinx.coroutines
Given Kotlin will support native code compilation in the future on everything from IOS, Android to the PC, the possibilities for having a single build which targets all platforms are getting closer.
H1B, the modern indentured servant
Didn't we outlaw slavery, indentured servants, etc. a while back?
Why hasn't Trump acted on his H1B promise?
https://www.theregister.co.uk/2017/03/13/h_1b_visa_reform_possibly_on_hold/
The IEEE is right. Large companies, some of which are hoarding massive amounts of cash, have been using the H1B to control tech wages. Some are the same companies which colluded w/ Steve Jobs illegal wage fixing pact.
https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_Litigation
https://www.theregister.co.uk/2015/04/21/apple_steve_jobs_poaching_lawsuit/
The settlement for the above is peanuts when compared to the actual damage done to tech wages. Frankly, there should be another class action lawsuit, targeting all companies which have abused the H1B, and covers every geek in the country.
The same companies are now using the H1B to control tech wages. This is wrong on so many levels. When will Trump raise the minimum H1B wage to 130k? Actually, in Silicon Valley, given the high cost of living, this should probably be something closer to 200k.
http://spectrum.ieee.org/view-from-the-valley/at-work/tech-careers/commentary-the-h1b-problem-as-ieeeusa-sees-it
Why hasn't Trump acted on his H1B promise?
https://www.theregister.co.uk/2017/03/13/h_1b_visa_reform_possibly_on_hold/
The IEEE is right. Large companies, some of which are hoarding massive amounts of cash, have been using the H1B to control tech wages. Some are the same companies which colluded w/ Steve Jobs illegal wage fixing pact.
https://en.wikipedia.org/wiki/High-Tech_Employee_Antitrust_Litigation
https://www.theregister.co.uk/2015/04/21/apple_steve_jobs_poaching_lawsuit/
The settlement for the above is peanuts when compared to the actual damage done to tech wages. Frankly, there should be another class action lawsuit, targeting all companies which have abused the H1B, and covers every geek in the country.
The same companies are now using the H1B to control tech wages. This is wrong on so many levels. When will Trump raise the minimum H1B wage to 130k? Actually, in Silicon Valley, given the high cost of living, this should probably be something closer to 200k.
http://spectrum.ieee.org/view-from-the-valley/at-work/tech-careers/commentary-the-h1b-problem-as-ieeeusa-sees-it
Monday, August 18, 2014
Saturday, June 21, 2014
Intel, Xeon, FPGA
I've been waiting years for an on die FPGA!
http://www.extremetech.com/extreme/184828-intel-unveils-new-xeon-chip-with-integrated-fpga-touts-20x-performance-boost
Time to brush up on the Verilog. Combine with C++ and mcorelab, we'll finally be able to really cook! High throughput message routing based on topic paths or tag/value pairs will now be possible. The line between network switch and application is finally starting to blur, and high throughput applications will be easier to build with commodity servers.
http://www.extremetech.com/extreme/184828-intel-unveils-new-xeon-chip-with-integrated-fpga-touts-20x-performance-boost
Time to brush up on the Verilog. Combine with C++ and mcorelab, we'll finally be able to really cook! High throughput message routing based on topic paths or tag/value pairs will now be possible. The line between network switch and application is finally starting to blur, and high throughput applications will be easier to build with commodity servers.
Thursday, June 12, 2014
Mcorelab - Mcoreweb bench numbers
Quick correction, numbers below are for Haswell with AVX2.
These are unofficial numbers, mileage may vary depending on your hardware. Still pretty amazing. With a low enough update rate I'd reckon one could get millions of websocket users on a single server!
http://www.mcorelab.com
test setup and some performance numbers of mcoreweb server (number reported are single server performance):
- server: dual E5-2690 with Intel 10GbE (X520).
- client: multiple client nodes generating load, total of 100,000 concurrent client sockets/connections.
- server answer each request message with a response message.
- message payload size=64Byte, so the test is the raw message/request handling capacity, rather than limited by PHY -- our software can easily saturate 10Gbit with just moderate message size)
Windows 2012/2008-R2:
1). sample server implemented on the mcoreweb/TCPserver layer:
using 2 core on the server: 1.5 Million/second,
using 4 core on the server: 3.3 Million/second
using 8 core on the server: 5.9 Million/second
SSL (AES encrypted):
using 4 core on the server: 1.35 Million/second
using 8 core on the server: 2 Million/second
(If using Intel newer gen Ivybridge CPU, SSL performance is much better. here the result is limited SandyBridge CPU which is without the new AVX2 instructions.)
2). sample server implemented on the mcoreweb/websocket layer:
2 core: 1.48 Million/second
4 core: 2.56 Million/second
8 core: 5.55 Million/second
Secure Websocket (AES encrypted):
4 core: 0.95 Million/second
8 core: 1.99 Million/second
Redhat Linux 6.4:
1). sample server implemented on the mcoreweb/TCPserver layer:
using 2 core on the server: 1.8 Million/second,
using 4 core on the server: 3.4 Million/second
using 8 core on the server: 6.1 Million/second
SSL (AES encrypted):
using 4 core on the server: 1.13 Million/second
using 8 core on the server: 2.1 Million/second
2). sample server implemented on the mcoreweb/websocket layer:
2 core: 1.43 Million/second
4 core: 3 Million/second
8 core: 5.45 Million/second
Secure Websocket (AES encrypted):
4 core: 0.99 Million/second
8 core: 1.75 Million/second
These are unofficial numbers, mileage may vary depending on your hardware. Still pretty amazing. With a low enough update rate I'd reckon one could get millions of websocket users on a single server!
http://www.mcorelab.com
test setup and some performance numbers of mcoreweb server (number reported are single server performance):
- server: dual E5-2690 with Intel 10GbE (X520).
- client: multiple client nodes generating load, total of 100,000 concurrent client sockets/connections.
- server answer each request message with a response message.
- message payload size=64Byte, so the test is the raw message/request handling capacity, rather than limited by PHY -- our software can easily saturate 10Gbit with just moderate message size)
Windows 2012/2008-R2:
1). sample server implemented on the mcoreweb/TCPserver layer:
using 2 core on the server: 1.5 Million/second,
using 4 core on the server: 3.3 Million/second
using 8 core on the server: 5.9 Million/second
SSL (AES encrypted):
using 4 core on the server: 1.35 Million/second
using 8 core on the server: 2 Million/second
(If using Intel newer gen Ivybridge CPU, SSL performance is much better. here the result is limited SandyBridge CPU which is without the new AVX2 instructions.)
2). sample server implemented on the mcoreweb/websocket layer:
2 core: 1.48 Million/second
4 core: 2.56 Million/second
8 core: 5.55 Million/second
Secure Websocket (AES encrypted):
4 core: 0.95 Million/second
8 core: 1.99 Million/second
Redhat Linux 6.4:
1). sample server implemented on the mcoreweb/TCPserver layer:
using 2 core on the server: 1.8 Million/second,
using 4 core on the server: 3.4 Million/second
using 8 core on the server: 6.1 Million/second
SSL (AES encrypted):
using 4 core on the server: 1.13 Million/second
using 8 core on the server: 2.1 Million/second
2). sample server implemented on the mcoreweb/websocket layer:
2 core: 1.43 Million/second
4 core: 3 Million/second
8 core: 5.45 Million/second
Secure Websocket (AES encrypted):
4 core: 0.99 Million/second
8 core: 1.75 Million/second
Subscribe to:
Posts (Atom)

