4 ms·
“In a few years, datacenter networks will grow an order of magnitude from 40Gb to 400Gb. Systems researchers, including myself, have been preparing for this new
by majke 4y ago
“In a few years, datacenter networks will grow an order of magnitude from 40Gb to 400Gb. Systems researchers, including myself, have been preparing for this new world for almost a decade [1,2]. Demikernel is the first fully-featured nanosecond-scale OS: 100ns to get a packet from the NIC to the app in Demikernel’s DPDK TCP stack, processing 1 million packets per second per core, etc [3].”
Meanwhile a typical server is doing couple of gbps effective bandwidth, running hundreds of interconnected internal services from redis via zookeeper, ceph, dozen of docker containers, prometheus, some virtualization (kvm?), dozen of sqlite instances, consul and many many many more small random things.
“Datacenter” i work with is not “datacenter” of doing 1mpps per core.
- jeffbee 4y agoAll progress happens at the limit. You may just be running a lot of crud that loafs along, but out there somewhere there is an operator of something that sends a million packets per second per core and they'd rather get 2 million. You don't use your SSD but there is some organization that finds 1 million IOPS to be a little lower than they'd like. You do benefit from these advances. Your searches get faster, your video streaming remains affordable, etc.
- crmd 4y agoHow do you find these limit users? My startup makes low cost block and file storage on AWS that does 2M IOPS per volume with 200 uSec latency, and we’re surprised to find most customers are perfectly happy using gp3 with 3k IOPs and 1-2 ms latency.
- daveguy 4y agoWhat's your startup? I'd be interested to see the cost. Persistence gotchas, encryption / security.
- bombcar 4y agoI have a suspicion that many of them are not running on AWS, since they'd like to remove as much cloud/network latency as possible.
- 8note 4y agoMy parents company is an example - they need that kind of performance in remote northern canada where connecting to AWS would be a couple bytes every couple seconds with a while of a latency. The server goes travelling instead, along with the people to use it
- oneplane 4y agoIf your application has to go to disk to fetch something it's already accepted that it'll take time. Things that need to be really fast go to RAM or cache. For everything else you're already getting into niches.
- rmah 4y agoHow can you drive by mentioning this without mentioning your company name?!?!? Our firm uses smallish sizes of AWS EFS storage and we're highly annoyed by the throughput limitations.
- 8note 4y agoThe limit work I've done is in geophysics, and AWS is a non-starter for data collection and analysis in the bush. If you can wait a few weeks to upload all the data to AWS, why are you going to care about moving the processing speed from a month to half a month? The high speed stuff is in dedicated hardware that can travel to where it's needed, whether that's in a base station or with the surveyors
- dilyevsky 4y agoI’d be interested in the company name too (feel free to dm if it’s stealth or something). I’d add that gp3 and other cloud pds implement ec under the hood and as such have very high durability and zero maintenance issues overhead (since they’re managed). And yeah you trade dogshit performance for that privilege