HNHacker News
TopNewBestAskShowJobs

ivankelly

67 karma · joined May 11, 2015

submissionscomments
ivankelly··on Cryptocurrency scam: Man swindled out of pension and life savings
name the celebrities at least.
ivankelly··on Celsius Appears insolvent, and it's taking the whole crypto market with it
gutness me, Ich needen ein stein of Doppelbock to get throu zis
ivankelly··on AWS costs every programmer should know (2019)
Inter-zone traffic is $0.01/GB EACH WAY. So in effect it's $0.02/GB https://www.duckbillgroup.com/blog/aws-cross-az-data-transfe...
ivankelly··on Ask HN: What game do you wish existed?
half-life 3
ivankelly··on With Proton and Steam Play, many Windows games now work on Linux
It's at the point now where the intro logos and stuff for games take longer than a reboot, so now I just reboot into windows.
ivankelly··on With Proton and Steam Play, many Windows games now work on Linux
Consider getting a laptop with TB3 and using an eGPU. I don't know how eGPUs work with linux though.
ivankelly··on EU drops rule that would prevent self-driving car data from being copyrighted
It would give you the unpersonalised experience then, where the car only knows how to drive in carparks in palo alto.
ivankelly··on EU drops rule that would prevent self-driving car data from being copyrighted
this is personal data, GDPR gives you the right to get a full dump of it.
ivankelly··on How did medieval people dance? (2013)
Sounds very like Sardana [1].

[1] https://en.wikipedia.org/wiki/Sardana

ivankelly··on Facebook open-sources LogDevice, a distributed storage for sequential data
What I was wondering really, was whether LogDevice provides total order atomic broadcast, and as such whether it solves concensus. It appears it does (or rather, it daisychains on the concensus provided by zookeeper and uses it's own fencing mechanism, similar to what bookkeeper/Pulsar does).
ivankelly··on Facebook open-sources LogDevice, a distributed storage for sequential data
Ok, but for this to be problematic, readers would need to have some other mechanism to know that w1 did actually take place before w2. So FIFO instead of total order.

Anyhow, thanks for answering my questions. Very interesting system.

ivankelly··on Facebook open-sources LogDevice, a distributed storage for sequential data
> Consistency in a more general sense than just read-modify-write consistency. If you have sequencers active in several epochs at the same time accepting writes, the records may end up being written out of order, and there would be a breakage of the total ordering guarantee.

But given that reads are blocked on all sequencers before the current one, this should still provide total order atomic broadcast, unless a single client can connect to a sequencer with a lower epoch than one it has already seen.

ivankelly··on Facebook open-sources LogDevice, a distributed storage for sequential data
> Yes, reads are not released (i.e. are blocked) until sealing is complete.

Which are the cases where consistency is compromised then? If a client of the log needs consistency, it needs to ensure that it has seen all previous updates to a log before making a new update, which implies a read.

> However, if your replication property, is say, "place 3 copies across 3 racks", then the definition of f-majority becomes more complicated - e.g. having all nodes in the nodeset respond minus two racks will also satisfy it.

Sure, the aim being that no write can be successfully acknowledged by enough replicas to complete the write.

ivankelly··on Facebook open-sources LogDevice, a distributed storage for sequential data
Does it block reads until sealing is complete? How many nodes in the nodeset have to respond before sealing is complete? [NodeSet] - [ReplicationFactor] + 1?
ivankelly··on Facebook open-sources LogDevice, a distributed storage for sequential data
What happens with a sequencer which appears to fail but hasn't really, and then comes back up after the process to figure out inflight records has completed? If that sequencer receives a record for a log, will it be able to write it to the storage nodes? I.e. is there any fencing mechanism to tell the storage nodes that the epoch has been bumped, so don't access writes for that epoch anymore?
← PreviousPage 2 of 2