From 0292d0f6ea96efc3af9195a249f6e9877c158769 Mon Sep 17 00:00:00 2001 From: David Smiley Date: Tue, 10 Apr 2018 23:20:12 -0400 Subject: [PATCH] SOLR-12160: Time Routed Alias (TRA) documentation --- solr/solr-ref-guide/src/collections-api.adoc | 39 ++++---- .../src/how-solrcloud-works.adoc | 3 +- .../src/time-routed-aliases.adoc | 95 +++++++++++++++++++ 3 files changed, 118 insertions(+), 19 deletions(-) create mode 100644 solr/solr-ref-guide/src/time-routed-aliases.adoc diff --git a/solr/solr-ref-guide/src/collections-api.adoc b/solr/solr-ref-guide/src/collections-api.adoc index 2f1ce2e6b96..1e895b287fb 100644 --- a/solr/solr-ref-guide/src/collections-api.adoc +++ b/solr/solr-ref-guide/src/collections-api.adoc @@ -480,26 +480,27 @@ http://localhost:8983/solr/admin/collections?action=DELETESHARD&collection=anoth [[createalias]] == CREATEALIAS: Create or Modify an Alias for a Collection -The `CREATEALIAS` action will create a new alias pointing to one or more collections. If an alias by the same name -already exists, this action will replace the existing alias, effectively acting like an atomic "MOVE" command. Aliases -come in 2 flavors: routed and non-routed. +The `CREATEALIAS` action will create a new alias pointing to one or more collections. +Aliases come in 2 flavors: standard and routed. -Non-routed aliases are simpler and can serve to rename a collection or to distribute queries across several collections. -While it is possible to send updates to an alias spanning multiple collections, non-routed alias have no logic for -distributing documents among the referenced collections so all updates will go to the first collection in the list. +*Standard aliases* are simple: CREATEALIAS registers the alias name with the names of one or more collections provided + by the command. +If an existing alias exists, it is replaced/updated. +A standard alias can serve to have the appearance of renaming a collection, and can be used to atomically swap +which backing/underlying collection is "live" for various purposes. +When Solr searches an alias pointing to multiple collections, Solr will search all shards of all the collections as an + aggregated whole. +While it is possible to send updates to an alias spanning multiple collections, standard aliases have no logic for + distributing documents among the referenced collections so all updates will go to the first collection in the list. `/admin/collections?action=CREATEALIAS&name=_name_&collections=_collectionlist_` -Routed aliases are more complex to configure, but include logic for automatically inspecting a field on the document -and using the value in that field to select a destination collection. Additionally, the routed alias automates the -partitioning of data across a series of collections by creating new collections periodically. This feature allows for -indefinite indexing of data without degradation of performance otherwise experienced due to the continuous growth of -an index. As new data arrives, a field on the document is inspected and the document is then potentially re-routed to -another collection. The underlying collections can can be queried independently but usually the alias will be used. -These collections are created automatically on the fly as new data arrives based on the parameters supplied in this -command. For very high volume use cases or for use cases requiring only a limited span of data to be retained, -collections older than a given age can be deleted. This delete of old collections only occurs if new documents are -sent to the alias. +*Routed aliases* are aliases with additional capabilities to act as a kind of super-collection -- routing + updates to the correct collection. +Since the only routing strategy at present is time oriented, these are also called *Time Routed Aliases* (TRAs). +A TRA manages an alias and a time sequential series of collections that it will both create and optionally delete on-demand. +See <> for some important high-level information + before getting started. NOTE: Presently this is only supported for temporal fields stored as a <> type. Other @@ -541,15 +542,17 @@ requirements for collection naming. `async`:: Request ID to track this action which will be <>. -==== Non-Routed Alias Parameters +==== Standard Alias Parameters `collections`:: A comma-separated list of collections to be aliased. The collections must already exist in the cluster. -This parameter signals the creation of a simple (non-routed) alias. If it is present all routing parameters are +This parameter signals the creation of a standard alias. If it is present all routing parameters are prohibited. If routing parameters are present this parameter is prohibited. ==== Routed Alias Parameters +Most routed alias parameters become _alias properties_ that can subsequently be inspected and <>. + `router.start`:: The start date/time of data for this time routed alias in Solr's standard date/time format (i.e., ISO-8601 or "NOW" optionally with <>). diff --git a/solr/solr-ref-guide/src/how-solrcloud-works.adoc b/solr/solr-ref-guide/src/how-solrcloud-works.adoc index fc68a70de9c..b0e42e42c06 100644 --- a/solr/solr-ref-guide/src/how-solrcloud-works.adoc +++ b/solr/solr-ref-guide/src/how-solrcloud-works.adoc @@ -1,5 +1,5 @@ = How SolrCloud Works -:page-children: shards-and-indexing-data-in-solrcloud, distributed-requests +:page-children: shards-and-indexing-data-in-solrcloud, distributed-requests, time-routed-aliases // Licensed to the Apache Software Foundation (ASF) under one // or more contributor license agreements. See the NOTICE file // distributed with this work for additional information @@ -21,6 +21,7 @@ The following sections cover provide general information about how various SolrC * <> * <> +* <> If you are already familiar with SolrCloud concepts and basic functionality, you can skip to the section covering <>. diff --git a/solr/solr-ref-guide/src/time-routed-aliases.adoc b/solr/solr-ref-guide/src/time-routed-aliases.adoc new file mode 100644 index 00000000000..7adf78398e0 --- /dev/null +++ b/solr/solr-ref-guide/src/time-routed-aliases.adoc @@ -0,0 +1,95 @@ += Time Routed Aliases + +Time Routed Aliases (TRAs) is a SolrCloud feature that manages an alias and a time sequential series of collections. +It automatically creates new collections and (optionally) deletes old ones as it routes documents to the correct + collection based on its timestamp. +This approach allows for indefinite indexing of data without degradation of performance otherwise experienced due to the + continuous growth of a single index. +If you need to store a lot of timestamped data in Solr, such as logs or IoT sensor data, then this feature probably + makes more sense than creating one sharded hash-routed collection. + +== How it works + +First you create a time routed aliases using the <> command with some + router settings. +Most of the settings are editable at a later time using the <> command. +The first collection will be created automatically, along with an alias pointing to it. +Each underlying Solr "core" in a collection that is a member of a TRA has a special core property referencing the alias. +The name of each collection is comprised of the TRA name and the start timestamp (UTC), with trailing zeros and symbols + truncated. +Ideally, as a user of this feature, you needn't concern yourself with the particulars of the collection naming pattern + since both queries and updates may be done via the alias. + +When adding data, you should usually direct documents to the alias (e.g. reference the alias name instead of any collection). +The Solr server and CloudSolrClient will direct an update request to the first collection that an alias points to. +The collections list for a TRA is always reverse sorted, and thus the connection path of the request will route to the + lead collection. Using CloudSolrClient is preferable as it can reduce the number of underlying physical HTTP requests by one. +If you know that a particular set of documents to be delivered is going to a particular older collection then you could + direct it there at the client side as an optimization but it's not necessary. CloudSolrClient does not (yet) do this. + +When processing an update for a TRA, Solr initializes its + <> chain as usual, but + when DistributedUpdateProcessor (DUP) initializes, it detects that the update targets a TRA and injects + TimeRoutedUpdateProcessor (TRUP) in front of itself. +TRUP, in coordination with the Overseer, is the main part of a TRA, and must immediately precede DUP. It is not + possible to configure custom chains with other types of UpdateRequestProcessors between TRUP and DUP. +TRUP first reads TRA configuration from the alias properties when it is initialized. As it sees each document, it checks for + changes to TRA properties, updates its cached configuration if needed and then determines which collection the + document belongs to: + +* If TRUP needs to send it to a time segment represented by a collection other than the one that + the client chose to communicate with, then it will do so using mechanisms shared with DUP. + Once the document is forwarded to the correct collection (i.e. the correct TRA time segment), it skips directly to + DUP on the target collection and continues normally, potentially being routed again to the correct shard & replica + within the target collection. + +* If it belongs in the current collection (which is usually the case if processing events as they occur), the document + passes through to DUP. DUP does it's normal collection-level processing that may involve routing the document + to another shard & replica. + +* If the time stamp on the document is more recent than the most recent TRA segment, then a new collection needs to be + added at the front of the TRA. + TRUP will create this collection, add it to the alias and then forward the document to the collection it just created. + This can happen recursively if more than one collection needs to be created. + Each time a new collection is added, the oldest collections in the TRA are examined for possible deletion, if that has + been configured. + All this happens synchronously, potentially adding seconds to the update request and indexing latency. + +Any other type of update like a commit or delete is routed by TRUP to all collections. +Generally speaking, this is not a performance concern. When Solr receives a delete or commit wherein nothing is deleted +or nothing needs to be committed, then it's pretty cheap. + +== Improvement Possibilities + +This is a new feature of SolrCloud that can be expected to be improved. +Some _potential_ areas for improvement that _are not implemented yet_ are: + +* Searches with time filters should only go to applicable collections. + +* Collections ought to be constrained by their size instead of or in addition to time. + Based on the underlying design, this would only apply to the lead collection. + +* Ways to automatically optimize (or reduce the resources of) older collections that aren't expected to receive more + updates, and might have less search demand. + +* New collections ought to be created preemptively, as an option, to avoid delaying a document that does not yet have + a collection to go to. + +* CloudSolrClient could route documents to the correct collection based on a timestamp instead always picking the + latest. + +== Limitations & Assumptions + +* Only *time* routed aliases are supported. If you instead have some other sequential number, you could fake it + as a time (e.g. convert to a timestamp assuming some epoch and increment). + The smallest possible interval is one second. + No other routing scheme is supported, although this feature was developed with considerations that it could be + extended/improved to other schemes. + +* The underlying collections form a contiguous sequence without gaps. This will not be suitable when there are + large gaps in the underlying data, as Solr will insist that there be a collection for each increment. This + is due in part on Solr calculating the end time of each interval collection based on the timestamp of + the next collection, since it is otherwise not stored in any way. + +* Avoid sending updates to the oldest collection if you have also configured that old collections should be + automatically deleted. It could lead to exceptions bubbling back to the indexing client. \ No newline at end of file