Proton is Vespa's search core. Proton maintains disk and memory structures for documents.
As the data is dynamic, these structures are periodically optimized by
maintenance jobs and the resource footprint of these background jobs
are primarily controlled by the
concurrency setting.
The content node has a bucket management system
which sends requests to a set of document databases,
which each consists of three sub-databasesready, not ready and removed:
Bucket management
When the node starts up it first needs to get an overview of what documents and hence buckets it has.
Once it knows this, it is in initializing mode, able to handle load,
but distributors do not yet know bucket metadata for all the buckets,
and thus can't know whether buckets are consistent with copies on other nodes.
Once metadata for all buckets are known, the content nodes transitions from
initializing to up state.
As the distributors wants quick access to bucket metadata,
it keeps an in-memory bucket database to efficiently serve these requests.
It implements elasticity support in terms of the SPI.
Operations are ordered according to priority,
and only one operation per bucket can be in-flight at a time.
Below bucket management is the persistence engine,
which implements the SPI in terms of Vespa search.
The persistence engine reads the document type from the document id,
and dispatches requests to the correct document database.
Document database
Each document database is responsible for a single document type.
It has a component called FeedHandler which takes care of incoming documents, updates, and remove requests.
All requests are first written to a transaction log,
then handed to the appropriate sub-database, based on the request type.
Sub-databases
There are three types of sub-databases, each with its own
document meta store and document store.
The document meta store holds a map from the document id to a local id.
This local id is used to address the document in the document store.
The document meta store also maintains information on the
state of the buckets that are present in the sub-database.
The sub-databases are maintained by the index maintainer.
The document distribution changes as the system is resized.
When the number of nodes in the system changes,
the index maintainer will move documents between the Ready and
Not Ready sub-databases to reflect the new distribution.
When an entry in the Removed sub-database gets old it is purged.
The sub-databases are:
Not Ready
Holds the redundant documents that are not searchable,
i.e. the not ready documents.
Documents that are not ready are only stored, not indexed.
It takes some processing to move from this state to the ready state.
Ready
Maintains an index of all ready documents and keeps them searchable.
One of the ready copies is active while the rest are not active:
Active
There should always be exactly one active copy of each
document in the system, though intermittently there may be more.
These documents produce results when queries are evaluated.
Not Active
The ready copies that are not active are indexed but will not produce results.
By being indexed, they are ready to take over immediately
if the node holding the active copy becomes unavailable.
Read more in searchable-copies.
Removed
Keeps track of documents that have been removed.
The id and timestamp for each document is kept.
This information is used when buckets from two nodes are merged.
If the removed document exists on another node but with a different timestamp,
the most recent entry prevails.
Transaction log
Content nodes have a transaction log to persist mutating operations.
The transaction log persists operations by file append.
Having a transaction log simplifies proton's in-memory index structures
and enables steady-state high performance, read more below.
All operations are written and synched to the transaction log.
This is sequential (not random) IO but might impact overall feed performance if running on NAS attached storage
where the synch operation has a much higher cost than on local attached storage (e.g. SSD).
See sync-transactionlog.
Default, proton will
flush components
like attribute vectors and memory index on shutdown, for quicker startup after scheduled restarts.
Index
Index fields are string fields, used for text search.
Other field types are attributes
and summary fields.
Proton stores position information in text indices by default, for proximity relevance -
posocc (below).
All the occurrences of a term is stored in the posting list, with its position.
This provides superior ranking features,
but is somewhat more expensive than just storing a single occurrence per document.
For most applications it is the correct tradeoff,
since most of the cost is usually elsewhere and relevance is valuable.
Applications that only need occurrence information for filtering
can use rank: filter
to optimize query performance, using only boolocc-files (below).
The memory index has a dictionary per index field.
This contains all unique words in that field with mapping to posting lists with position information.
The position information is used during ranking, see
nativeRank for details on how a text match score is calculated.
The disk index stores the content of each index field in separate folders.
Each folder contains:
Compressed posting lists with position information. File: posocc.dat.compressed.
Posting lists with only occurrence information (bitvector). These are generated for common words.
Files: boolocc.bdat, boolocc.idx.
Example:
$ pwd
/opt/vespa/var/db/vespa/search/cluster.mycluster/n1/documents/myschema/0.ready/index/index.flush.1/myfield
$ ls-la
total 7632
drwxr-xr-x 2 org users 145 Oct 29 06:09 .
drwxr-xr-x 74 org users 4096 Oct 29 06:11 ..
-rw-r--r-- 1 org users 4096 Oct 29 06:11 boolocc.bdat
-rw-r--r-- 1 org users 4096 Oct 29 06:11 boolocc.idx
-rw-r--r-- 1 org users 8192 Oct 29 06:11 dictionary.pdat
-rw-r--r-- 1 org users 8192 Oct 29 06:11 dictionary.spdat
-rw-r--r-- 1 org users 4120 Oct 29 06:11 dictionary.ssdat
-rw-r--r-- 1 org users 7778304 Oct 29 06:11 posocc.dat.compressed
Note that boolocc-files are empty if number of occurrences is small, like in the example above.
Document store
Documents are stored in the document store.
Put, update and remove operations are persisted in the transaction log
before updating the document in the document store.
The operation is ack'ed to the client and the result of the operation is immediately seen in search results.
Files in the document store are written sequentially, and occur in pairs - example:
-rw-r--r-- 1 owner users 4133380096 Aug 10 13:36 1467957947689211000.dat
-rw-r--r-- 1 owner users 71192112 Aug 10 13:36 1467957947689211000.idx
The maximum size:
(in bytes) per .dat file on disk can be set using the following:
The files are written in sequence. proton starts with one pair
and grows it until maxfilesize. Once full, a new pair is started.
This means, the pair is immutable, except the last pair, which is written to.
Documents exist in multiple versions in multiple files.
Older versions are compacted away when a pair is scheduled for being the new active pair -
obsolete versions are removed, leaving only the active document version left in a new file pair
- which is the new active pair.
Files are written in chunks,
using compression settings.
Defragmentation
Document store compaction,
defragments and sort document store files.
It removes stale versions of documents (i.e. old version of updated documents).
It is triggered when the disk bloat of the document store is larger than the total disk usage of the document store
times
diskbloatfactor.
Refer to summary tuning for details.
Defragmentation status is hence best observed by tracking max_bucket_spread over time,
a sawtooth pattern is normal for corpora that change over time.
The document_store_compact metric tracks when proton is running compaction jobs.
Compaction settings can be set too tight, in that case, the metric is always, or close to, 1.
When benchmarking, it is hence important to set the correct compaction settings,
and also make sure that proton has compacted files since (can take hours),
and is not actively compacting (document_store_compact should be 0 most of the time).
Note:
There is no bucket-compaction across files - documents will not move between files.
Optimized reads using chunks
As documents are clustered within the .dat file,
proton optimizes reads by reading larger chunks when accessing documents.
When visiting, documents are read in bucket order.
This is the same order as the defragmentation jobs uses.
The first document read in a visit operation for a bucket will read a chunk from the .dat file into memory.
Subsequent document accesses are served be a memory lookup only.
The chunk size is configured by
maxsize:
There can be 2^22=4M chunks. This sets a minimum chunk size based on maxfilesize -
e.g. an 8G file can have minimum 2k chunk size.
Finally, bucket size is configured by setting
bucket-splitting:
The following are hence the relevant sizing units:
.dat file size - maxfilesize.
Larger files give less files and hence better locality,
but compaction requires more memory and more time to complete.
chunk size - maxsize.
Smaller chunks give less wasted IO bytes but more IO operations.
bucket size - bucket-splitting.
Larger buckets give less buckets and hence better locality to nodes and files.
Memory usage
The document store has a mapping in memory from local ID (LID) to position in a document store file (.dat).
Part of this mapping is persisted in the .idx-file paired to the .dat file.
The memory used by the document store is linear with number of documents and updates to these.
The metric content.proton.documentdb.ready.document_store.memory_usage.allocated_bytes
gives the size in memory -
use the metric API to find it.
A rule of thumb is 12 bytes per document.
Proton maintenance jobs
Tune the jobs in Proton tuning.
Sizing search describes the static proton sizing -
this article details the temporary resource usage for the proton jobs.
There is only one instance of each job at a time - e.g. attributes are flushed in sequence.
When a job is running, its metric is set to 1 - otherwise 0.
Use this to correlate observed performance with job runs - see Run metric.
Flush an attribute vector from memory to disk,
based on configuration in the
flushstrategy.
This controls memory usage and query performance.
This also makes proton starts quicker - see
flush on shutdown.
CPU
Little - one thread flushes to disk
Memory
Little - some temporary use
Disk
A new file is written to, hence 2x the size of an attribute on disk.
Flush a memory index to disk, then trigger disk index fusion.
The goal is to shrink memory usage by adding to the disk-backed indices.
Performance characteristics for this flush is similar to indexing.
Note: A high feed rate can cause multiple smaller flushed indices, like
$VESPA_HOME/var/db/vespa/search/cluster.name/n1/documents/doc/0.ready/index/index.flush.102
- see the high index number.
Multiple smaller indices is a symptom of too small memory indices compared to feed rate -
to fix, increase
flushstrategy > native > component > maxmemorygain.
CPU
Little - one thread indexes to disk
Memory
Little
Disk
Creates a new disk index, size of the memory index.
Defragment and sort document store files as documents are updated and deleted,
in order to reduce disk usage.
The file is sorted in bucket order on output. Triggered by
diskbloatfactor.
CPU
Little - one thread reads one files, sorts and writes a new file
Memory
Holds a document summary store file in memory plus memory for sorting the file.
Note: This is important on hosts with little memory!
Reduce
maxfilesize to increase number of files and use less temporary memory for compaction.
Disk
A new file is written while the current is serving, max temporary usage is 2x.
Triggered by nodes going up/down, refer to
content cluster elasticity and
searchable-copies.
It causes documents to be indexed or de-indexed, similar to feeding.
This moves documents in or out of ready/active sub-databases.
CPU
CPU similar to feeding.
Consumes capacity from the index write thread - hence has feeding impact
Memory
As feeding - the memory index will grow
Disk
As feeding
Run metric
content.proton.documentdb.job.bucket_move
lid-space compaction
As bucket move, however moves documents within a sub-database.
This is often triggered when a cluster grows with more nodes,
documents are redistributed to new nodes and each node has fewer documents -
a LIDspace compaction is hence triggered. This in-place defragments the
document meta store.
Resources are freed on a subsequent attribute flush.
Retrieving documents is done by specifying an id to get,
or use a selection expression
to visit a range of documents - refer to the Document API.
Overview:
Get
When the content node receives a get request,
it scans through all the document databases,
and for each one it checks all three sub-databases.
Once the document is found, the scan is stopped and the document returned.
If the document is found in a Ready sub-database,
the document retriever will apply any changes that is stored in the
attributes before returning the document.
Visit
A visit request creates an iterator over each candidate bucket.
This iterator will retrieve matching documents from all sub-databases of all document databases.
As for get, attributes values are applied to document fields in the Ready sub-database.
Queries
Queries has a separate pathway through the system.
It does not use the distributor, nor has it anything to do with the SPI.
It is orthogonal to the elasticity set up by the storage and retrieval described above.
How queries move through the system:
A query enters the system through the QR-server (query rewrite server)
in the Vespa Container.
The QR-server issues one query per document type to the search nodes:
Container
The Container knows all the document types and rewrites queries as a
collection of queries, one for each type.
Queries may have a restrict parameter,
in which case the container will send the query only to the specified document types.
It sends the query to content nodes and collect partial results.
It pings all content nodes every second to know whether they are alive,
and keeps open TCP connections to each one.
If a node goes down, the elastic system will make the documents available on other nodes.
Content node matching
The match engine
receives queries and routes them to the right document database based on the document type.
The query is passed to the Ready sub-database, where the searchable documents are.
Based on information stored in the document meta store,
the query is augmented with a blocklist that ensures only active documents are matched.
Custom Component State API
This section describes the custom extensions of the proton custom component state API.
Component status can be found by HTTP GET at http://host:stateport/state/v1/custom/component.
This gives an overview of the relevant search node components and their internal state.
Note that this is not a stable API, and it will expand and change between releases.
