[
https://issues.apache.org/jira/browse/CASSANDRA-9666?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15110894#comment-15110894
]
Jeff Jirsa commented on CASSANDRA-9666:
---------------------------------------
My understanding, following CASSANDRA-10280 and others, is that DTCS is exactly
where I said it would be in July of last year:
bq. Instead, you'd have some new parameter that says "I want the largest window
no larger than X" (seconds? days? megabytes?), below which you'd have a tiering
system that does weird things (by our experience, and I imagine by yours, too,
as you note DTCS gets confused with invalid timestamps in CASSANDRA-8340), and
above which is essentially a fixed window.
bq. Don't we end up with something that looks very much like TWCS, with a
difficult to tune, confusion prone tiering below some cutoff parameter?
We're at that point - the two primary differences are:
- minTimestamp vs maxTimestamp, which is trivial to change, but is the primary
cause of the write amplification for most real world clusters (read repair
happens), and
- Weird parameters in DTCS. Expecting real users to do math with
base_time_second is not realistic. I couldn't calculate the impact and I was
knee deep in the code. When dealing with hundreds of terabytes, having exact
parameters I can reason about costs or saves me real money when we talk about
efficient expiration of data.
The weird tiering in DTCS that may help some people in some cases, but I don't
know who would want it or in what cases it's desirable.
DTCS is close enough to TWCS that we don't need both in the tree. The beautiful
thing about Cassandra is that it's pluggable. I'm fine maintaining TWCS outside
the tree.
> Provide an alternative to DTCS
> ------------------------------
>
> Key: CASSANDRA-9666
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9666
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Jeff Jirsa
> Assignee: Jeff Jirsa
> Fix For: 2.1.x, 2.2.x
>
> Attachments: dtcs-twcs-io.png, dtcs-twcs-load.png
>
>
> DTCS is great for time series data, but it comes with caveats that make it
> difficult to use in production (typical operator behaviors such as bootstrap,
> removenode, and repair have MAJOR caveats as they relate to
> max_sstable_age_days, and hints/read repair break the selection algorithm).
> I'm proposing an alternative, TimeWindowCompactionStrategy, that sacrifices
> the tiered nature of DTCS in order to address some of DTCS' operational
> shortcomings. I believe it is necessary to propose an alternative rather than
> simply adjusting DTCS, because it fundamentally removes the tiered nature in
> order to remove the parameter max_sstable_age_days - the result is very very
> different, even if it is heavily inspired by DTCS.
> Specifically, rather than creating a number of windows of ever increasing
> sizes, this strategy allows an operator to choose the window size, compact
> with STCS within the first window of that size, and aggressive compact down
> to a single sstable once that window is no longer current. The window size is
> a combination of unit (minutes, hours, days) and size (1, etc), such that an
> operator can expect all data using a block of that size to be compacted
> together (that is, if your unit is hours, and size is 6, you will create
> roughly 4 sstables per day, each one containing roughly 6 hours of data).
> The result addresses a number of the problems with
> DateTieredCompactionStrategy:
> - At the present time, DTCS’s first window is compacted using an unusual
> selection criteria, which prefers files with earlier timestamps, but ignores
> sizes. In TimeWindowCompactionStrategy, the first window data will be
> compacted with the well tested, fast, reliable STCS. All STCS options can be
> passed to TimeWindowCompactionStrategy to configure the first window’s
> compaction behavior.
> - HintedHandoff may put old data in new sstables, but it will have little
> impact other than slightly reduced efficiency (sstables will cover a wider
> range, but the old timestamps will not impact sstable selection criteria
> during compaction)
> - ReadRepair may put old data in new sstables, but it will have little impact
> other than slightly reduced efficiency (sstables will cover a wider range,
> but the old timestamps will not impact sstable selection criteria during
> compaction)
> - Small, old sstables resulting from streams of any kind will be swiftly and
> aggressively compacted with the other sstables matching their similar
> maxTimestamp, without causing sstables in neighboring windows to grow in size.
> - The configuration options are explicit and straightforward - the tuning
> parameters leave little room for error. The window is set in common, easily
> understandable terms such as “12 hours”, “1 Day”, “30 days”. The
> minute/hour/day options are granular enough for users keeping data for hours,
> and users keeping data for years.
> - There is no explicitly configurable max sstable age, though sstables will
> naturally stop compacting once new data is written in that window.
> - Streaming operations can create sstables with old timestamps, and they'll
> naturally be joined together with sstables in the same time bucket. This is
> true for bootstrap/repair/sstableloader/removenode.
> - It remains true that if old data and new data is written into the memtable
> at the same time, the resulting sstables will be treated as if they were new
> sstables, however, that no longer negatively impacts the compaction
> strategy’s selection criteria for older windows.
> Patch provided for :
> - 2.1: https://github.com/jeffjirsa/cassandra/commits/twcs-2.1
> - 2.2: https://github.com/jeffjirsa/cassandra/commits/twcs-2.2
> - trunk (post-8099): https://github.com/jeffjirsa/cassandra/commits/twcs
> Rebased, force-pushed July 18, with bug fixes for estimated pending
> compactions and potential starvation if more than min_threshold tables
> existed in current window but STCS did not consider them viable candidates
> Rebased, force-pushed Aug 20 to bring in relevant logic from CASSANDRA-9882
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
