[
https://issues.apache.org/jira/browse/HBASE-14070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16096927#comment-16096927
]
Appy commented on HBASE-14070:
------------------------------
PT = physical time, LT = logical time, ST = system time
----
Note that in current implementation, master and RSs are updating their own
clocks on receiving any region close/open request/response.
Also, on receiving a clock ahead of its own, they update their own clock to its
PT+LT, and keep increasing LT till their own ST catches that PT.
----
Problem 1: cascading logical time increment
When more RS are involved say - 3 RS and 1 master. Let's say max skew is 30
sec.
HLC Clocks (physical time, logical time): X = don't care
RS1: (50, 100k)
Master: (40, X)
RS2: (30, X)
RS3: (20, X)
[RS3's ST behind RS1's by 30 sec.]
RS1 replies to master, sends it's clock (50,X).
Master's clock (50, X). It'll be another 10 sec before it's own physical clock
reaches 50, so HLC's PT will remain 50 for next 10 sec.
Master --> RS2
RS2's clock = (50, X).
RS2 keeps incrementing LT on writes (since it's own PT is behind) for few
seconds before it replies back to master with (50, X+ few 100k).
Master's clock = (50, X+ few 100k) [Since master's physical clock hasn't caught
up yet, note that it was 10 seconds behind, PT remains 50.].
Master --> RS3
RS3's clock (50, X+few 100k)
But RS3's ST is behind RS1's ST by 30 sec, which means it'll keep incrementing
LT for next 30 sec (unless it gets a newer clock from master).
But the problem is, RS3 has much smaller LT window than actual 1M!!
---
Problem 2:
Single bad RS clock crashing the cluster:
If a single RS's clock is bad and a bit faster, it'll catch time and keep
pulling master's PT with it. If 'real time' is say 20, max skew time is 10, and
bad RS is at time 29.9, it'll pull master to 29.9 (via next response), and then
any RS less than 19.9, i.e. just 0.1 sec away from real time will die due to
higher than max skew.
This can bring whole clusters down!
---
Problem 3: Time jumps (not a bug, but more of a nuisance)
Say a RS is behind master by 20 sec. On each communication from master, RS will
update its own PT to master's PT, and it'll remain that till RS's ST catches
up. If there are frequent communication from master, ST might never catch up
and RS's PT will actually look like discrete time jumps rather than continuous
time.
For eg. If master communicated with RS at times 30, 40, 50 (RSs corresponding
times are 10, 20, 30), than all events on RS between time [10, 50] will be
timestamped with either 30, 40 or 50.
---
> Hybrid Logical Clocks for HBase
> -------------------------------
>
> Key: HBASE-14070
> URL: https://issues.apache.org/jira/browse/HBASE-14070
> Project: HBase
> Issue Type: New Feature
> Reporter: Enis Soztutar
> Assignee: Amit Patel
> Attachments: HBASE-14070.master.001.patch,
> HybridLogicalClocksforHBaseandPhoenix.docx,
> HybridLogicalClocksforHBaseandPhoenix.pdf
>
>
> HBase and Phoenix uses systems physical clock (PT) to give timestamps to
> events (read and writes). This works mostly when the system clock is strictly
> monotonically increasing and there is no cross-dependency between servers
> clocks. However we know that leap seconds, general clock skew and clock drift
> are in fact real.
> This jira proposes using Hybrid Logical Clocks (HLC) as an implementation of
> hybrid physical clock + a logical clock. HLC is best of both worlds where it
> keeps causality relationship similar to logical clocks, but still is
> compatible with NTP based physical system clock. HLC can be represented in
> 64bits.
> A design document is attached and also can be found here:
> https://docs.google.com/document/d/1LL2GAodiYi0waBz5ODGL4LDT4e_bXy8P9h6kWC05Bhw/edit#
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
