Kirill,
Thanks for driving this. This is awaited by many users.
A few comments and questions.
I would keep CacheWarmup interface purely internal and never view it as an
interface which a user would be implementing.
There are multiple reasons for that:
- The logic of the cache warmup is very low-level; how a user is supposed to
know which pages they want?
- A sophisticated strategy will require accessing private APIs for sure; say, I
need a strategy which loads the last known memory state before the restart; how
can I even implement that without breaking into various internals?
- In fact there aren't many implementations which make sense ("load
everything", "load indexes", "load last memory state", "load N GB at random");
every use case I've seen would be solved by a "load everything" strategy (if
disk is < RAM) or "load last memory state" strategy
- Warmup will be a critical phase, and a custom user implementation is all too
likely to cause issues. We should avoid executing user code in critical stages
if we can help it
To summarize, if we give warmup strategies in users' hands they will be hard to
write, will require breaking into internals or a lot of additional public
interfaces for these internals, will likely cause issues with the cluster, and
everyone will be implementing the same few general strategies.
Basically, I expect only fellow Ignite developers to be implementing their own
strategies.
Because of that I propose to keep the interfaces private, and only give a
single public parameter. The parameter can take an enum of the supported
strategies. New useful strategies should be added to Ignite codebase.
Will there be a way to interrupt warmup phase and continue startup (e.g. via
JMX, REST and/or control.sh)? Can we have it please?
I think that ideally warmup should be configured per-cache - I believe this is
what a user would expect to do.
However, cache configs are immutable. We need a way for existing users to enjoy
the cache warmup feature, as well as for early adopters to switch to more
sophisticated strategies as they will be released (or as their dataset grows).
Because of that I propose to add the cache warmup configuration to the
DataRegionConfiguration. Data regions can be changed between restarts,
independently on each node allowing for a rolling change.
Will preloadPartition() method be deprecated together with this change? I
assume yes?
How hard would it be to implement a "load all indexes, metapages and freelists"
strategy in addition to the "load everything"?
I think it would be an MVP for environments with a datasets larger than RAM. A
"load everything" strategy will not work in this environments pretty much at
all, and "load indexes" will be a significant improvement to no warmup at all.
Thanks,
Stan
> On 4 Aug 2020, at 16:04, ткаленко кирилл <tkalkir...@yandex.ru> wrote:
>
> Hi, Denis!
>
> For now, I suggest a simple warm-up implementation, if the persistent storage
> is less than RAM. If others want to make additional implementations, they can
> do it themselves by implementing interfaces. For the first point, we need to
> figure out how and where we will remember pages, etc. Perhaps for such tasks
> it will be necessary to make improvements in kernel.
>
> In "WarmUpStrategy#warmUp" method, we get "GridKernalContext#cache" from
> which we can get with caches and groups through
> "GridCacheProcessor#cacheGroups", "GridCacheProcessor#caches" and so on, we
> can access to pages.
>> The second one requires direct work with data pages, but not with a cache
>> context, so it's also impossible to implement.
>
> This requires writing additional custom code, which may run longer due to its
> SQL features, and so on.
> It would be more convenient to just set a warm-up strategy for both developer
> and grid administrator.
>> When loading of all cache data is required, it can be done by running a
>> local scan query. It will iterate through all data pages and result in
>> their allocation in memory.
>
> 04.08.2020, 15:25, "Denis Mekhanikov" <dmekhani...@gmail.com>:
>> Kirill,
>>
>> When I discussed this functionality with Ignite users, I heard the
>> following thoughts about warming up:
>>
>> - Node restarts affect performance of queries. The main reason for that
>> is that the pages that were loaded into memory before the restart are on
>> disk after the restart. It takes time to reach the same distribution of
>> data between memory and disk. Until that point the performance is usually
>> degraded. No simple rule like "load everything" helps here if only a part
>> of data fits in memory.
>> - It would be nice to have a way to give preferences to indices when
>> doing a warmup. Usually indices are used more often than data nodes, so
>> loading indices first would bring more benefits.
>>
>> The first point can be addressed by implementing the policy that would
>> restore the memory state that was observed before the restart. I don't see
>> how it can be implemented using the suggested interface.
>> The second one requires direct work with data pages, but not with a cache
>> context, so it's also impossible to implement.
>>
>> When loading of all cache data is required, it can be done by running a
>> local scan query. It will iterate through all data pages and result in
>> their allocation in memory.
>>
>> So, I don't really see a scenario when the suggested API will help. Do you
>> have a suitable use-case that will be covered?
>>
>> Denis
>>
>> вт, 4 авг. 2020 г. в 13:42, ткаленко кирилл <tkalkir...@yandex.ru>:
>>
>>> Hi, Denis!
>>>
>>> Previously, I answered Slava about implementation that I keep in mind, now
>>> it will be possible to add own warm-up strategy implementations. Which will
>>> be possible to implement in different ways.
>>>
>>> At the moment, I suggest implementing one "Load all" strategy, which will
>>> be effective if persistent storage is less than RAM.
>>>
>>> 28.07.2020, 19:46, "Denis Mekhanikov" <dmekhani...@gmail.com>:
>>> > Kirill,
>>> >
>>> > That will be a great feature! Other popular databases already have it
>>> (e.g.
>>> > Postgres: https://www.postgresql.org/docs/11/pgprewarm.html), so it's
>>> good
>>> > that we're also going to have it in Ignite.
>>> >
>>> > What implementation of CacheWarmup interface do you have in mind? Will
>>> > there be some preconfigured implementation, and will users be able to
>>> > implement it themselves?
>>> >
>>> > Do you think it should be cache-based? I would say that a
>>> DataRegion-based
>>> > warm-up would come more naturally. Page IDs that are loaded into the data
>>> > region can be dumped periodically to disk and recovered on restarts. This
>>> > is more or less how it works in Postgres.
>>> > I'm afraid that if we make it cache-based, the implementation won't be
>>> that
>>> > obvious. We already have an API for warmup that appeared to be pretty
>>> much
>>> > impossible to apply in a useful way:
>>> >
>>>
>>> https://ignite.apache.org/releases/latest/javadoc/org/apache/ignite/IgniteCache.html#preloadPartition-int-
>>> > Let's make sure that our new tool for warming up is actually useful.
>>> >
>>> > Denis
>>> >
>>> > вт, 28 июл. 2020 г. в 09:17, Zhenya Stanilovsky
>>> <arzamas...@mail.ru.invalid
>>> >> :
>>> >
>>> >> Looks like we need additional func for static caches, for
>>> >> example: warmup(List<CacheConfiguration> cconf) it would be helpful for
>>> >> spring too.
>>> >>
>>> >> >
>>> >> >------- Forwarded message -------
>>> >> >From: "Вячеслав Коптилин" < slava.kopti...@gmail.com >
>>> >> >To: dev@ignite.apache.org
>>> >> >Cc:
>>> >> >Subject: Re: [DISCUSSION] Cache warmup
>>> >> >Date: Mon, 27 Jul 2020 16:47:48 +0300
>>> >> >
>>> >> >Hello Kirill,
>>> >> >
>>> >> >Thanks a lot for driving this activity. If I am not mistaken, this
>>> >> >discussion relates to IEP-40.
>>> >> >
>>> >> >> I suggest adding a warmup phase after recovery here [1] after [2],
>>> >> before
>>> >> >discovery.
>>> >> >This means that the user's thread, which starts Ignite via
>>> >> >Ignition.start(), will wait for ana additional step - cache warm-up.
>>> >> >I think this fact has to be clearly mentioned in our documentation (at
>>> >> >Javadocat least) because this step can be time-consuming.
>>> >> >
>>> >> >> I suggest adding a new interface:
>>> >> >I would change it a bit. First of all, it would be nice to place this
>>> >> >interface to a public package and get rid of using GridCacheContext,
>>> >> >which is an internal class and it should not leak to the public API
>>> in any
>>> >> >case.
>>> >> >Perhaps, this parameter is not needed at all or we should add some
>>> public
>>> >> >abstraction instead of internal class.
>>> >> >
>>> >> >package org.apache.ignite.configuration;
>>> >> >
>>> >> >import org.apache.ignite.IgniteCheckedException;
>>> >> >import org.apache.ignite.lang.IgniteFuture;
>>> >> >
>>> >> >public interface CacheWarmupper {
>>> >> > /**
>>> >> > * Warmup cache.
>>> >> > *
>>> >> > * @param cachename Cache name.
>>> >> > * @return Future cache warmup.
>>> >> > * @throws IgniteCheckedException If failed.
>>> >> > */
>>> >> > IgniteFuture<?> warmup(String cachename) throws
>>> >> >IgniteCheckedException;
>>> >> >}
>>> >> >
>>> >> >Thanks,
>>> >> >S.
>>> >> >
>>> >> >пн, 27 июл. 2020 г. в 15:03, ткаленко кирилл < tkalkir...@yandex.ru
>>> >:
>>> >> >
>>> >> >> Now, after restarting node, we have only cold caches, which at first
>>> >> >> requests to them will gradually load data from disks, which can slow
>>> >> down
>>> >> >> first calls to them.
>>> >> >> If node has more RAM than data on disk, then they can be loaded at
>>> start
>>> >> >> "warmup", thereby solving the issue of slowdowns during first calls
>>> to
>>> >> >> caches.
>>> >> >>
>>> >> >> I suggest adding a warmup phase after recovery here [1] after [2],
>>> >> before
>>> >> >> descovery.
>>> >> >>
>>> >> >> I suggest adding a new interface:
>>> >> >>
>>> >> >> package org.apache.ignite.internal.processors.cache;
>>> >> >>
>>> >> >> import org.apache.ignite.IgniteCheckedException;
>>> >> >> import org.apache.ignite.internal.IgniteInternalFuture;
>>> >> >> import org.jetbrains.annotations.Nullable;
>>> >> >>
>>> >> >> /**
>>> >> >> * Interface for warming up cache.
>>> >> >> */
>>> >> >> public interface CacheWarmup {
>>> >> >> /**
>>> >> >> * Warmup cache.
>>> >> >> *
>>> >> >> * @param cacheCtx Cache context.
>>> >> >> * @return Future cache warmup.
>>> >> >> * @throws IgniteCheckedException if failed.
>>> >> >> */
>>> >> >> @Nullable IgniteInternalFuture<?> process(GridCacheContext cacheCtx)
>>> >> >> throws IgniteCheckedException;
>>> >> >> }
>>> >> >>
>>> >> >> Which will allow to warm up caches in parallel and asynchronously.
>>> >> Warmup
>>> >> >> phase will end after all IgniteInternalFuture for all caches isDone.
>>> >> >>
>>> >> >> Also adding the ability to customize via methods:
>>> >> >>
>>> >>
>>> org.apache.ignite.configuration.IgniteConfiguration#setDefaultCacheWarmup
>>> >> >> org.apache.ignite.configuration.CacheConfiguration#setCacheWarmup
>>> >> >>
>>> >> >> Which will allow for each cache to set implementation of cache
>>> warming
>>> >> >> up,
>>> >> >> both for a specific cache, and for all if necessary.
>>> >> >>
>>> >> >> I suggest adding an implementation of SequentialWarmup that will use
>>> >> [3].
>>> >> >>
>>> >> >> Questions, suggestions, comments?
>>> >> >>
>>> >> >> [1] -
>>> >> >>
>>> >>
>>>
>>> org.apache.ignite.internal.processors.cache.GridCacheProcessor.CacheRecoveryLifecycle#afterLogicalUpdatesApplied
>>> >> >> [2] -
>>> >> >>
>>> >>
>>>
>>> org.apache.ignite.internal.processors.cache.GridCacheProcessor.CacheRecoveryLifecycle#restorePartitionStates
>>> >> >> [3] -
>>> >> >>
>>> >>
>>>
>>> org.apache.ignite.internal.processors.cache.IgniteCacheOffheapManager.CacheDataStore#preload
