4/ Problems (3): - disk throughput can be throttled (GP2, IO1) - if you use SSDs on cloud, which model do you have, are you the only writer?, how to get a proper RAID setup?, ... - how to minimize the number of switches between your instances? - unpredictable network behaviors
-
Show this thread
-
5/ Problems (4): - reliable UDP multicast: AWS seems not to support UDP multicast, while GC does not. Anyway what's the expected packet loss? - security: spectre attacks, data hosted on 3rd party, ... -
#crypto is growing. When next bull run? How to ensure 10x/100x scalability2 replies 1 retweet 17 likesShow this thread -
6/ Path to metal


Our infra team has outstanding experience in secure, god wrath resistant, deployments, so we discussed a lot about pros and cons.
Here are the 2 cons summarized:
- hardware decay
- physical hosts maintenance (trips to datacenter, hardware expertise, ...)3 replies 2 retweets 17 likesShow this thread -
7/ Benefit (1): - the entire infra behavior is predictable - RAID enabled historical sharded data-warehouses kick asses - KYD, KYM and KYC
: Know Your Disk, Know your Mem and Know Your CPU
- dedicated switches and firewalls, back to back connectivity for core servers1 reply 1 retweet 25 likesShow this thread -
8/ Benefit (2): - reliable UDP multicast - bare-metal Linux installs - you're the only one using your CPU (see spectre-like attacks) - colocation services offering
2 replies 1 retweet 15 likesShow this thread -
-
-
Replying to @paoloardoino
Do you happen to know an open source trading engine that you’d consider a good starting point to experiment with?
1 reply 0 retweets 0 likes -
Replying to @alpha_authority
There are many on GitHub (search open source matching engine) Ideally a trading engine is as simple as 2 binary trees (bid, ask), 1 input queue and 1 output queue. Managing complex order types, make it extremely performant and reliable is the complex part.
1 reply 0 retweets 2 likes -
Replying to @paoloardoino
Thanks, for me performance is even more important than reliability since I want to do some faster than realtime simulations. Do you have experience whether Node makes sense at all or better do the extra hop to C++ even though it is supposed to run as an API service eventually?
1 reply 0 retweets 0 likes
I would suggest to go with C++ or Java. Node garbage collection is not as tweakable as Java (ZGC is really cool).
-
-
Replying to @paoloardoino
Been ages since I used C++ but was one of my first languages to earn - kind of excited… hehe
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.