Co-Founder & CSO @brainportsai | Technical Leader | Cloud & AI Infrastructure Architect | Previously @intel, @ONF_SDN, @OracleDataSci, etc.
California
Joined July 2012
- Tweets2.6K
- Following523
- Followers384
- Likes2.2K
Jordan Halterman ✌️ retweeted
Transactions that are fast, simple and fault tolerant! 😲 exciting work by @ChrisJe34211511
Wait… what!?
Fault tolerant 2PC that is simple and commits in 1RTT? 🤯
If you missed @ChrisJe34211511 fantastic talk in #eurosys24 (#papoc24) definitely check the paper bit.ly/U2PCp
Jordan Halterman ✌️ retweeted
When I first saw @TigerBeetleDB's Simulator and how every possible failure is enumerated and subsequently tested for, I was instantly hooked
Model checking the actual system 🏴☠️
github.com/tigerbeetle/tiger…
Jordan Halterman ✌️ retweeted
Types of Distributed Systems Papers.
Joke modeled after @xkcd 's xkcd.com/2456/
#distributedsystems #distributedsystemsjokes
It seems sharing it was worthwhile! I think SDN-based approaches to consensus are really promising, and with some time and effort we could see similar techniques brought to real world systems.
I wish I had the opportunity to work on consensus algorithms more often. I miss it 😌
I was cleaning out my GitHub repos the other day and came across this old gem I’d totally forgotten about. When I was at the Open Networking Foundation, I did some work researching low latency consensus using SDN-enabled clock synchronization protocols.
github.com/kuujo/just-in-tim…
I was cleaning out my GitHub repos the other day and came across this old gem I’d totally forgotten about. When I was at the Open Networking Foundation, I did some work researching low latency consensus using SDN-enabled clock synchronization protocols.
github.com/kuujo/just-in-tim…
The JIT Paxos protocol itself is largely s derived from on Viewstamped Replication (leader, views, etc). Requests are sent by the client to all replicas, and consensus is achieved in a single round trip as long as messages arrive in wall clock order.
I’d certainly expect its performance to degrade significantly under high load at least. (although hopefully no more than a traditional consensus algorithm). It’s clearly not ready for the real world. But I thought it would be interesting to share nonetheless… for posterity.
Yes. If a new leader is elected for any reason, that leader can overwrite the entry with an entry from its term. An entry is not guaranteed to be retained until a leader commits it (commit index >= entry index). 1/
Raft Consensus Challenge II ⛓️
We have a Raft cluster with 3 nodes, each maintaining a replica of the log. Node 2 is the leader of term 3, accepted client request 9, & added 9 to its log.
Can 9 be lost? Why?
teapaw.org/DominikTornow/status/1…
Prior to committing an entry from its term, entries in a leader’s term can be overwritten even if they’re stored on a majority of replicas. See figure 8 in the original Raft paper for a deep dive into how and why that can happen:
raft.github.io/raft.pdf
3/
I think the answer I’m expected to give is node 2… but I’m going to “well, actually” this and point out it could be node 1 OR 2. IIRC there’s nothing in the Raft protocol that says a leader must append to its log before sending AppendEntries RPCs to followers.
1/
Raft Consensus Challenge ⛓️
We have a Raft cluster with 3 nodes, each maintaining a replica of the log.
Which node is the leader for term 3?! What gives it away?!
#ThinkingInDistributedSystems #ConsensusChallenge
But of course, this will only be the case if the leader has committed entires in its term. Until the leader commits an entry from its term (to make a majority of the logs consistent at the start of the term) the commit index will have been set by a prior leader. 4/4