gvsafronov/futriix
1
0
Fork 0
Code Issues Pull Requests Actions 4 Packages Projects Releases Wiki Activity
futriix/tests/unit/dump.tcl
uriyage bbfd041895
Async IO threads (#758) 
This PR is 1 of 3 PRs intended to achieve the goal of 1 million requests
per second, as detailed by [dan touitou](https://github.com/touitou-dan)
in https://github.com/valkey-io/valkey/issues/22. This PR modifies the
IO threads to be fully asynchronous, which is a first and necessary step
to allow more work offloading and better utilization of the IO threads.

### Current IO threads state:

Valkey IO threads were introduced in Redis 6.0 to allow better
utilization of multi-core machines. Before this, Redis was
single-threaded and could only use one CPU core for network and command
processing. The introduction of IO threads helps in offloading the IO
operations to multiple threads.

**Current IO Threads flow:**

1. Initialization: When Redis starts, it initializes a specified number
of IO threads. These threads are in addition to the main thread, each
thread starts with an empty list, the main thread will populate that
list in each event-loop with pending-read-clients or
pending-write-clients.
2. Read Phase: The main thread accepts incoming connections and reads
requests from clients. The reading of requests are offloaded to IO
threads. The main thread puts the clients ready-to-read in a list and
set the global io_threads_op to IO_THREADS_OP_READ, the IO threads pick
the clients up, perform the read operation and parse the first incoming
command.
3. Command Processing: After reading the requests, command processing is
still single-threaded and handled by the main thread.
4. Write Phase: Similar to the read phase, the write phase is also be
offloaded to IO threads. The main thread prepares the response in the
clients’ output buffer then the main thread puts the client in the list,
and sets the global io_threads_op to the IO_THREADS_OP_WRITE. The IO
threads then pick the clients up and perform the write operation to send
the responses back to clients.
5. Synchronization: The main-thread communicate with the threads on how
many jobs left per each thread with atomic counter. The main-thread
doesn’t access the clients while being handled by the IO threads.

**Issues with current implementation:**

* Underutilized Cores: The current implementation of IO-threads leads to
the underutilization of CPU cores.
* The main thread remains responsible for a significant portion of
IO-related tasks that could be offloaded to IO-threads.
* When the main-thread is processing client’s commands, the IO threads
are idle for a considerable amount of time.
* Notably, the main thread's performance during the IO-related tasks is
constrained by the speed of the slowest IO-thread.
* Limited Offloading: Currently, Since the Main-threads waits
synchronously for the IO threads, the Threads perform only read-parse,
and write operations, with parsing done only for the first command. If
the threads can do work asynchronously we may offload more work to the
threads reducing the load from the main-thread.
* TLS: Currently, we don't support IO threads with TLS (where offloading
IO would be more beneficial) since TLS read/write operations are not
thread-safe with the current implementation.

### Suggested change

Non-blocking main thread - The main thread and IO threads will operate
in parallel to maximize efficiency. The main thread will not be blocked
by IO operations. It will continue to process commands independently of
the IO thread's activities.

**Implementation details**

**Inter-thread communication.**

* We use a static, lock-free ring buffer of fixed size (2048 jobs) for
the main thread to send jobs and for the IO to receive them. If the ring
buffer fills up, the main thread will handle the task itself, acting as
back pressure (in case IO operations are more expensive than command
processing). A static ring buffer is a better candidate than a dynamic
job queue as it eliminates the need for allocation/freeing per job.
* An IO job will be in the format: ` [void* function-call-back | void
*data] `where data is either a client to read/write from and the
function-ptr is the function to be called with the data for example
readQueryFromClient using this format we can use it later to offload
other types of works to the IO threads.
* The Ring buffer is one way from the main-thread to the IO thread, Upon
read/write event the main thread will send a read/write job then in
before sleep it will iterate over the pending read/write clients to
checking for each client if the IO threads has already finished handling
it. The IO thread signals it has finished handling a client read/write
by toggling an atomic flag read_state / write_state on the client
struct.

**Thread Safety**

As suggested in this solution, the IO threads are reading from and
writing to the clients' buffers while the main thread may access those
clients.
We must ensure no race conditions or unsafe access occurs while keeping
the Valkey code simple and lock free.

Minimal Action in the IO Threads
The main change is to limit the IO thread operations to the bare
minimum. The IO thread will access only the client's struct and only the
necessary fields in this struct.
The IO threads will be responsible for the following:

* Read Operation: The IO thread will only read and parse a single
command. It will not update the server stats, handle read errors, or
parsing errors. These tasks will be taken care of by the main thread.
* Write Operation: The IO thread will only write the available data. It
will not free the client's replies, handle write errors, or update the
server statistics.


To achieve this without code duplication, the read/write code has been
refactored into smaller, independent components:

* Functions that perform only the read/parse/write calls.
* Functions that handle the read/parse/write results.

This refactor accounts for the majority of the modifications in this PR.

**Client Struct Safe Access**

As we ensure that the IO threads access memory only within the client
struct, we need to ensure thread safety only for the client's struct's
shared fields.

* Query Buffer 
* Command parsing - The main thread will not try to parse a command from
the query buffer when a client is offloaded to the IO thread.
* Client's memory checks in client-cron - The main thread will not
access the client query buffer if it is offloaded and will handle the
querybuf grow/shrink when the client is back.
* CLIENT LIST command - The main thread will busy-wait for the IO thread
to finish handling the client, falling back to the current behavior
where the main thread waits for the IO thread to finish their
processing.
* Output Buffer 
* The IO thread will not change the client's bufpos and won't free the
client's reply lists. These actions will be done by the main thread on
the client's return from the IO thread.
* bufpos / block→used: As the main thread may change the bufpos, the
reply-block→used, or add/delete blocks to the reply list while the IO
thread writes, we add two fields to the client struct: io_last_bufpos
and io_last_reply_block. The IO thread will write until the
io_last_bufpos, which was set by the main-thread before sending the
client to the IO thread. If more data has been added to the cob in
between, it will be written in the next write-job. In addition, the main
thread will not trim or merge reply blocks while the client is
offloaded.
* Parsing Fields 
    * Client's cmd, argc, argv, reqtype, etc., are set during parsing.
* The main thread will indicate to the IO thread not to parse a cmd if
the client is not reset. In this case, the IO thread will only read from
the network and won't attempt to parse a new command.
* The main thread won't access the c→cmd/c→argv in the CLIENT LIST
command as stated before it will busy wait for the IO threads.
* Client Flags 
* c→flags, which may be changed by the main thread in multiple places,
won't be accessed by the IO thread. Instead, the main thread will set
the c→io_flags with the information necessary for the IO thread to know
the client's state.
* Client Close 
* On freeClient, the main thread will busy wait for the IO thread to
finish processing the client's read/write before proceeding to free the
client.
* Client's Memory Limits 
* The IO thread won't handle the qb/cob limits. In case a client crosses
the qb limit, the IO thread will stop reading for it, letting the main
thread know that the client crossed the limit.

**TLS**

TLS is currently not supported with IO threads for the following
reasons:

1. Pending reads - If SSL has pending data that has already been read
from the socket, there is a risk of not calling the read handler again.
To handle this, a list is used to hold the pending clients. With IO
threads, multiple threads can access the list concurrently.
2. Event loop modification - Currently, the TLS code
registers/unregisters the file descriptor from the event loop depending
on the read/write results. With IO threads, multiple threads can modify
the event loop struct simultaneously.
3. The same client can be sent to 2 different threads concurrently
(https://github.com/redis/redis/issues/12540).

Those issues were handled in the current PR:

1. The IO thread only performs the read operation. The main thread will
check for pending reads after the client returns from the IO thread and
will be the only one to access the pending list.
2. The registering/unregistering of events will be similarly postponed
and handled by the main thread only.
3. Each client is being sent to the same dedicated thread (c→id %
num_of_threads).


**Sending Replies Immediately with IO threads.**

Currently, after processing a command, we add the client to the
pending_writes_list. Only after processing all the clients do we send
all the replies. Since the IO threads are now working asynchronously, we
can send the reply immediately after processing the client’s requests,
reducing the command latency. However, if we are using AOF=always, we
must wait for the AOF buffer to be written, in which case we revert to
the current behavior.

**IO threads dynamic adjustment**

Currently, we use an all-or-nothing approach when activating the IO
threads. The current logic is as follows: if the number of pending write
clients is greater than twice the number of threads (including the main
thread), we enable all threads; otherwise, we enable none. For example,
if 8 IO threads are defined, we enable all 8 threads if there are 16
pending clients; else, we enable none.
It makes more sense to enable partial activation of the IO threads. If
we have 10 pending clients, we will enable 5 threads, and so on. This
approach allows for a more granular and efficient allocation of
resources based on the current workload.

In addition, the user will now be able to change the number of I/O
threads at runtime. For example, when decreasing the number of threads
from 4 to 2, threads 3 and 4 will be closed after flushing their job
queues.

**Tests**

Currently, we run the io-threads tests with 4 IO threads
(443d80f168/.github/workflows/daily.yml (L353)).
This means that we will not activate the IO threads unless there are 8
(threads * 2) pending write clients per single loop, which is unlikely
to happened in most of tests, meaning the IO threads are not currently
being tested.

To enforce the main thread to always offload work to the IO threads,
regardless of the number of pending events, we add an
events-per-io-thread configuration with a default value of 2. When set
to 0, this configuration will force the main thread to always offload
work to the IO threads.

When we offload every single read/write operation to the IO threads, the
IO-threads are running with 100% CPU when running multiple tests
concurrently some tests fail as a result of larger than expected command
latencies. To address this issue, we have to add some after or wait_for
calls to some of the tests to ensure they pass with IO threads as well.

Signed-off-by: Uri Yagelnik <uriy@amazon.com>
2024-07-08 20:01:39 -07:00

412 lines
15 KiB
Tcl
Raw Blame History

start_server {tags {"dump"}} {
test {DUMP / RESTORE are able to serialize / unserialize a simple key} {
r set foo bar
set encoded [r dump foo]
r del foo
list [r exists foo] [r restore foo 0 $encoded] [r ttl foo] [r get foo]
} {0 OK -1 bar}
test {RESTORE can set an arbitrary expire to the materialized key} {
r set foo bar
set encoded [r dump foo]
r del foo
r restore foo 5000 $encoded
set ttl [r pttl foo]
assert_range $ttl 3000 5000
r get foo
} {bar}
test {RESTORE can set an expire that overflows a 32 bit integer} {
r set foo bar
set encoded [r dump foo]
r del foo
r restore foo 2569591501 $encoded
set ttl [r pttl foo]
assert_range $ttl (2569591501-3000) 2569591501
r get foo
} {bar}
test {RESTORE can set an absolute expire} {
r set foo bar
set encoded [r dump foo]
r del foo
set now [clock milliseconds]
r restore foo [expr $now+3000] $encoded absttl
set ttl [r pttl foo]
assert_range $ttl 2000 3100
r get foo
} {bar}
test {RESTORE with ABSTTL in the past} {
r set foo bar
set encoded [r dump foo]
set now [clock milliseconds]
r debug set-active-expire 0
r restore foo [expr $now-3000] $encoded absttl REPLACE
catch {r debug object foo} e
r debug set-active-expire 1
set e
} {ERR no such key} {needs:debug}
test {RESTORE can set LRU} {
r set foo bar
set encoded [r dump foo]
r del foo
r config set maxmemory-policy allkeys-lru
r restore foo 0 $encoded idletime 1000
set idle [r object idletime foo]
assert {$idle >= 1000 && $idle <= 1010}
assert_equal [r get foo] {bar}
r config set maxmemory-policy noeviction
} {OK} {needs:config-maxmemory}
test {RESTORE can set LFU} {
r set foo bar
set encoded [r dump foo]
r del foo
r config set maxmemory-policy allkeys-lfu
r restore foo 0 $encoded freq 100
set freq [r object freq foo]
assert {$freq == 100}
r get foo
assert_equal [r get foo] {bar}
r config set maxmemory-policy noeviction
} {OK} {needs:config-maxmemory}
test {RESTORE returns an error of the key already exists} {
r set foo bar
set e {}
catch {r restore foo 0 "..."} e
set e
} {*BUSYKEY*}
test {RESTORE can overwrite an existing key with REPLACE} {
r set foo bar1
set encoded1 [r dump foo]
r set foo bar2
set encoded2 [r dump foo]
r del foo
r restore foo 0 $encoded1
r restore foo 0 $encoded2 replace
r get foo
} {bar2}
test {RESTORE can detect a syntax error for unrecognized options} {
catch {r restore foo 0 "..." invalid-option} e
set e
} {*syntax*}
test {RESTORE should not store key that are already expired, with REPLACE will propagate it as DEL or UNLINK} {
r del key1{t} key2{t}
r set key1{t} value2
r lpush key2{t} 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
r set key{t} value
set encoded [r dump key{t}]
set now [clock milliseconds]
set repl [attach_to_replication_stream]
# Keys that have expired will not be stored.
r config set lazyfree-lazy-server-del no
assert_equal {OK} [r restore key1{t} [expr $now-5000] $encoded replace absttl]
r config set lazyfree-lazy-server-del yes
assert_equal {OK} [r restore key2{t} [expr $now-5000] $encoded replace absttl]
assert_equal {0} [r exists key1{t} key2{t}]
# Verify the propagate of DEL and UNLINK.
assert_replication_stream $repl {
{select *}
{del key1{t}}
{unlink key2{t}}
}
close_replication_stream $repl
} {} {needs:repl}
test {DUMP of non existing key returns nil} {
r dump nonexisting_key
} {}
test {MIGRATE is caching connections} {
# Note, we run this as first test so that the connection cache
# is empty.
set first [srv 0 client]
r set key "Some Value"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert_match {*migrate_cached_sockets:0*} [r -1 info]
r -1 migrate $second_host $second_port key 9 1000
assert_match {*migrate_cached_sockets:1*} [r -1 info]
}
} {} {external:skip}
test {MIGRATE cached connections are released after some time} {
after 15000
assert_match {*migrate_cached_sockets:0*} [r info]
}
test {MIGRATE is able to migrate a key between two instances} {
set first [srv 0 client]
r set key "Some Value"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists key] == 1}
assert {[$second exists key] == 0}
set ret [r -1 migrate $second_host $second_port key 9 5000]
assert {$ret eq {OK}}
assert {[$first exists key] == 0}
assert {[$second exists key] == 1}
assert {[$second get key] eq {Some Value}}
assert {[$second ttl key] == -1}
}
} {} {external:skip}
test {MIGRATE is able to copy a key between two instances} {
set first [srv 0 client]
r del list
r lpush list a b c d
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists list] == 1}
assert {[$second exists list] == 0}
set ret [r -1 migrate $second_host $second_port list 9 5000 copy]
assert {$ret eq {OK}}
assert {[$first exists list] == 1}
assert {[$second exists list] == 1}
assert {[$first lrange list 0 -1] eq [$second lrange list 0 -1]}
}
} {} {external:skip}
test {MIGRATE will not overwrite existing keys, unless REPLACE is used} {
set first [srv 0 client]
r del list
r lpush list a b c d
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists list] == 1}
assert {[$second exists list] == 0}
$second set list somevalue
catch {r -1 migrate $second_host $second_port list 9 5000 copy} e
assert_match {ERR*} $e
set ret [r -1 migrate $second_host $second_port list 9 5000 copy replace]
assert {$ret eq {OK}}
assert {[$first exists list] == 1}
assert {[$second exists list] == 1}
assert {[$first lrange list 0 -1] eq [$second lrange list 0 -1]}
}
} {} {external:skip}
test {MIGRATE propagates TTL correctly} {
set first [srv 0 client]
r set key "Some Value"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists key] == 1}
assert {[$second exists key] == 0}
$first expire key 10
set ret [r -1 migrate $second_host $second_port key 9 5000]
assert {$ret eq {OK}}
assert {[$first exists key] == 0}
assert {[$second exists key] == 1}
assert {[$second get key] eq {Some Value}}
assert {[$second ttl key] >= 7 && [$second ttl key] <= 10}
}
} {} {external:skip}
test {MIGRATE can correctly transfer large values} {
set first [srv 0 client]
r del key
for {set j 0} {$j < 40000} {incr j} {
r rpush key 1 2 3 4 5 6 7 8 9 10
r rpush key "item 1" "item 2" "item 3" "item 4" "item 5" \
"item 6" "item 7" "item 8" "item 9" "item 10"
}
assert {[string length [r dump key]] > (1024*64)}
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists key] == 1}
assert {[$second exists key] == 0}
set ret [r -1 migrate $second_host $second_port key 9 10000]
assert {$ret eq {OK}}
assert {[$first exists key] == 0}
assert {[$second exists key] == 1}
assert {[$second ttl key] == -1}
assert {[$second llen key] == 40000*20}
}
} {} {external:skip}
test {MIGRATE can correctly transfer hashes} {
set first [srv 0 client]
r del key
r hmset key field1 "item 1" field2 "item 2" field3 "item 3" \
field4 "item 4" field5 "item 5" field6 "item 6"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists key] == 1}
assert {[$second exists key] == 0}
set ret [r -1 migrate $second_host $second_port key 9 10000]
assert {$ret eq {OK}}
assert {[$first exists key] == 0}
assert {[$second exists key] == 1}
assert {[$second ttl key] == -1}
}
} {} {external:skip}
test {MIGRATE timeout actually works} {
set first [srv 0 client]
r set key "Some Value"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists key] == 1}
assert {[$second exists key] == 0}
set rd [valkey_deferring_client]
$rd debug sleep 1.0 ; # Make second server unable to reply.
after 100; # wait to make sure DEBUG command was executed.
set e {}
catch {r -1 migrate $second_host $second_port key 9 500} e
assert_match {IOERR*} $e
}
} {} {external:skip}
test {MIGRATE can migrate multiple keys at once} {
set first [srv 0 client]
r set key1 "v1"
r set key2 "v2"
r set key3 "v3"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
assert {[$first exists key1] == 1}
assert {[$second exists key1] == 0}
set ret [r -1 migrate $second_host $second_port "" 9 5000 keys key1 key2 key3]
assert {$ret eq {OK}}
assert {[$first exists key1] == 0}
assert {[$first exists key2] == 0}
assert {[$first exists key3] == 0}
assert {[$second get key1] eq {v1}}
assert {[$second get key2] eq {v2}}
assert {[$second get key3] eq {v3}}
}
} {} {external:skip}
test {MIGRATE with multiple keys must have empty key arg} {
catch {r MIGRATE 127.0.0.1 6379 NotEmpty 9 5000 keys a b c} e
set e
} {*empty string*} {external:skip}
test {MIGRATE with multiple keys migrate just existing ones} {
set first [srv 0 client]
r set key1 "v1"
r set key2 "v2"
r set key3 "v3"
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
set ret [r -1 migrate $second_host $second_port "" 9 5000 keys nokey-1 nokey-2 nokey-2]
assert {$ret eq {NOKEY}}
assert {[$first exists key1] == 1}
assert {[$second exists key1] == 0}
set ret [r -1 migrate $second_host $second_port "" 9 5000 keys nokey-1 key1 nokey-2 key2 nokey-3 key3]
assert {$ret eq {OK}}
assert {[$first exists key1] == 0}
assert {[$first exists key2] == 0}
assert {[$first exists key3] == 0}
assert {[$second get key1] eq {v1}}
assert {[$second get key2] eq {v2}}
assert {[$second get key3] eq {v3}}
}
} {} {external:skip}
test {MIGRATE with multiple keys: stress command rewriting} {
set first [srv 0 client]
r flushdb
r mset a 1 b 2 c 3 d 4 c 5 e 6 f 7 g 8 h 9 i 10 l 11 m 12 n 13 o 14 p 15 q 16
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
set ret [r -1 migrate $second_host $second_port "" 9 5000 keys a b c d e f g h i l m n o p q]
assert {[$first dbsize] == 0}
assert {[$second dbsize] == 15}
}
} {} {external:skip}
test {MIGRATE with multiple keys: delete just ack keys} {
set first [srv 0 client]
r flushdb
r mset a 1 b 2 c 3 d 4 c 5 e 6 f 7 g 8 h 9 i 10 l 11 m 12 n 13 o 14 p 15 q 16
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
$second mset c _ d _; # Two busy keys and no REPLACE used
catch {r -1 migrate $second_host $second_port "" 9 5000 keys a b c d e f g h i l m n o p q} e
assert {[$first dbsize] == 2}
assert {[$second dbsize] == 15}
assert {[$first exists c] == 1}
assert {[$first exists d] == 1}
}
} {} {external:skip}
test {MIGRATE AUTH: correct and wrong password cases} {
set first [srv 0 client]
r del list
r lpush list a b c d
start_server {tags {"repl"}} {
set second [srv 0 client]
set second_host [srv 0 host]
set second_port [srv 0 port]
$second config set requirepass foobar
$second auth foobar
assert {[$first exists list] == 1}
assert {[$second exists list] == 0}
set ret [r -1 migrate $second_host $second_port list 9 5000 AUTH foobar]
assert {$ret eq {OK}}
assert {[$second exists list] == 1}
assert {[$second lrange list 0 -1] eq {d c b a}}
r -1 lpush list a b c d
$second config set requirepass foobar2
catch {r -1 migrate $second_host $second_port list 9 5000 AUTH foobar} err
assert_match {*WRONGPASS*} $err
}
} {} {external:skip}
}
Reference in New Issue View Git Blame Copy Permalink