| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
|
|
|
|
| |
The board is Vega's examples/shared-board.aup, and Vega's shared_board
check runs its two nodes: changes on either agree, and a partition
healed agrees on the later change. Reading a named field list once a
delivery would save part of 310 ns a record during upgrades alone: not
worth an API. What review 004 asked that is left is 103's third item,
broader performance comparisons under conditions stated first.
|
| |
|
|
|
|
|
|
|
| |
aup spreads an item or a message into a send or a record, and sets an
item whole (c69682e): the board sends { ..at } and applies cards[k] =
Card { ..remote }, and a new field is written in two places, the record
and the output. v6 on both nodes, cut apart and joined again, agrees on
the later change; its module is byte for byte the one with every field
written.
|
| |
|
|
|
|
|
| |
A board record of five fields (38 B) sent with v3's output fields took
310 ns to lay out for v4's input, reading the field list each time.
Only streams of another shape pay it, and only while nodes are upgraded
one after the other: less than a delivery's own work.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
103's next step: grow a real app, adding and reordering fields, cutting
the connection, starting again, and write down what gets in the way.
Two nodes share a board (from Vega's examples/board.aup), each sending
its changes to the other, the later change winning.
Upgrading them one after the other deadlocked: fixed by recasting
waiting sends by name (9aa123e) and taking fields by name (014,
5ce7655), and under it a link that stopped for good when a refused
stream's try was lost (baaf955). Now the older node's changes reach the
newer one during the upgrade, the newer one's new field waits for the
older, and the boards agree a second after. Fields in another order
agree at once; a partition healed agrees on the later change; both
nodes started again go on. Two rough edges eased on the way (the node's
name, message like in aup); sends and applies are still written field
by field. CHECKPOINT's row for links says so.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
013's second step is in (d9e0b68): each transport holds a window of its
output's sends. 104 measures it with the receivers down: once the window
is full a node grows by the outbox alone, 219 and 218 B a send over the
link and HTTP (399 and 534 B before).
The third step is a measurement. A checkpoint holds one list's snapshot
while it writes the image, so what it adds is what the app changed
meanwhile: nothing visible on tmpfs or for a list that fits the card's
fast writes; on the SD card, a 22.4 MB list written for about 8 s while
1,400 items a second were set left +3.0 and +4.4 MiB. The highest
resident memory was twice the state bytes, inside 013's bound (the state
and one snapshot) and its stop criterion (3x).
021's run.py takes --items, --stores (a directory on the device to
measure) and --label, and reports the peak (VmHWM).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
104 measures a node's memory under a receiver that is down and under a
snapshot held for long (021's run.py: resident memory against what was
posted, 30 s a case). A held snapshot grew the node without end (6.4 to
678 MiB, fixed in d1e8f38); a slow receiver grows it by what was sent,
221 B a send, which is state until delivered, so the budget bounds state
and refuses past it.
013 is that budget's design: what it counts (state bytes, deterministic),
what an event over it does, why an answer takes its send's room, how
links and HTTP wait, what it does not count and those bounds, and the
checks. With --memory 4000000 the same load fills it in about 4 s and
the node stays at 9.6-9.8 MiB (45.9 MiB and growing without).
012 records the hole the budget found in the link (the full flag), and
103 that its clock fix fell short (385754c). CHECKPOINT gains the row.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Three broke what data means (a walk's range missed the indices an insert
moves; an uncertain save of numbers alone let the next save through; an
output whose fields swapped names kept its waiting sends), four were
errors (a stateless timer never ticked, Vega's terminal lost a key cut
across reads, an input named edit forced documents, timers waited on the
wall clock). Each is fixed with a check that fails on the code before
and passes now, in Pure's and Vega's make check. The review's other
advice (a whole app's memory budget, one real app grown, comparisons with
their conditions set first) is what comes next.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
102: the spec (012) and a receiver written from it alone; the four holes
writing it found (REFUSED, the record's size, a peer's run places, the
memo's places beside) and the credit M6 showed fixed at 1,024 records;
M6 with ngtcp2 as a third, two rounds. Under reordering Pure carries the
most (45 Mbit/s; ngtcp2 29-47, quinn 1-2), under random loss the three
are the same, on a clean bottleneck Pure carries 3-5 % less (its framing,
some 112 B a packet to QUIC's 41), and where loss and reordering meet
ngtcp2 carries 1.3-3.6 times as much: undoing a loss it took for
reordering, it restores the window a real loss had cut too (its log: 22
of 34 undos past the window before the last cut; none without
reordering). Shares beside TCP stay within 25 %.
011 and the status table say so; 012 also says that a receiver answers
within 25 ms (a sender's PTO counts on it), that a shape and a record's
bytes are the app's, and why a packet of the past does no harm.
|
| |
|
|
|
|
|
|
|
| |
A path MTU black hole is seen and the link goes back to 1,200 bytes; a
receiver that lost its memo says LOST and M2 without a memo catches up
in 0.4-0.5 s, not 7.7; M5 stands beside ngtcp2 too (Pure 25 % lighter
than both QUIC from 1 KiB, 1.04 of ngtcp2 and 0.90 of quinn at 61 B);
and the stream's name in every packet, 096's guess for 61 B's gap, cost
nothing measurable, so it stays. 098 notes the one line LOST changes.
|
| |
|
|
|
|
|
|
|
| |
(096, 100)
61-byte records: 0.287 us a record against quinn's 0.31 in the same five
rounds (0.93 of it); 096's 1.06-1.12 was against quinn at 0.27, which
moves between 0.27 and 0.32 from day to day. H4 is no longer at the
edge.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
H3 of design 011 stands: with ten thousand devices sending 2,000 records
a second, the node's store, key and memo carried to another address,
nothing stopped for a 100 ms step, the moved node made no X25519, and
every record was applied once. Each device's first record after the move
waits one probe (30-45 ms here), as its first packet finds no way there.
The search found the listener answering from an address other than the
one a packet came to, and a link that looked for its receiver after its
probe went, losing every answer. H1-H4 are all measured now; 61 B's CPU
a delivery is at the edge of the criterion, to be looked at before step
5 (adoption and a specification).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Forward secrecy, 098's first task left: keys that leak later open only
epoch 0. M2 after the epoch began and after a restart is as before it
(p50 0.4 ms, p99 0.8 ms, the same CPU, no X25519 on restart). It costs a
p50 of 0.7-1.0 ms more while ten thousand devices first meet and for
about 40 s of accepts after, and a receiver that lost its memo now
stalls 0.8 s and catches up in 7.7 s, not 0.2-0.7: it lost the epochs,
and each sender opens one again. The search also found a packet more a
device from each of M6's path probe and the security checks' challenge,
and a load generator whose own pass over its devices was the floor of
the latency it measured. 098's note that epochs narrow KCI to the first
epoch was wrong; corrected.
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
(097)
M6 of design 011: with a 50 Mbit/s bottleneck, a link carries 21-41
times quinn's goodput when packets reorder, the same under random loss,
and 1.3-3.5 times with both; its p99 at 50 records a second is the same
or shorter; beside TCP CUBIC it takes 0.48-0.55 of the path, inside the
guardrail. Only on a clean path is it 3-5% below quinn.
|
| |
|
|
|
| |
libpure-link.a holds them; pure-serve and the M5 harness link it instead
of naming its source files.
|
| |
|
|
|
|
|
| |
Between two pure-serve nodes, 16 KiB records drain at 33,600 a second
(23,000 before), the links spending 58 us a record (76), under quinn's 60
while the receiver also stores each one. 1,013 bytes: 406,000 a second
(347,000). Saving sends one by one: 5,000 of 16 KiB in 0.91 s (84.8).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
M5 of design 011. At first a link spent eight times quinn's CPU on a small
record: scans of the backlog, a system call and allocations a record, 19
bytes of framing. After the seven changes before this one, the transport
alone (transport.c; a backlog drained, every record checked) spends per
record 0.30 us for 61 bytes (quinn 0.27: 1.06 to 1.12 times over three
campaigns, at the design's 10 % line), 2.91 us for 1,013 bytes (quinn
3.88) and 45.2 us for 16,013 bytes (quinn 60.9): 25 % less for the larger
ones, and 30 % more records a second. What is left for small records is
per packet: the stream's names and the RECORDS head. quinn's numbers carry
no confirmation of storing and no exactly once.
Between two pure-serve nodes the links spend more per record, woken at
the receiving app's pace. Filling the sender showed the outbox is written
whole at every save, as the numbers heard were (094): the next thing to
change.
run.py gains --transport and --drain; the quinn service reads a drain a
chunk at a time and its client writes 64 KiB at a time.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
M2 and M4 of design 011. Ten thousand devices send an order every five
seconds (2,000 a second); the receiver is killed (-9) at about 21 s and
started again at once. With its secrets remembered the node shows no
100 ms step without an order stored, catches up in 0.2 s, spends no more
CPU than before and makes no X25519; 34 orders were sent again, and
every order made was applied once (three rounds; on the SD card the
same, with 0.2 s silent). Without the memo it makes 10,000 X25519, 1.6 s
of CPU. quinn under the same load stops for 1.0 s and catches up after
10.9 s: 10,115 new connections, about 7 s more of the service's CPU and
8 s of the clients'. A first contact takes Pure one round trip and QUIC
two, at 10, 50 and 100 ms of RTT.
run.py --restart and --first-contact make them; the quinn harness gains
a fixed port with a reset key from its key, clients that connect again,
and a drain for M5.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
M1 of design 011. A Pure node allows 1,000 to 1,000,000 senders; each
sends one order and, once told it is stored, rests. At rest a link costs
the node 132 bytes (the sender allowed, 73; its number heard, about 60)
and no transport state; with a million the node runs in 139 MB. A quinn
0.11 service holding the same exchange's clients connected at rest costs
38,106 bytes a connection: 289 times as much, past the design's 1/100.
A link at rest takes its next record in 0.56 ms to stored, one X25519 on
the node included; a new QUIC connection takes 1.06 ms and 2.8 times the
receiver's CPU, without storing.
research/experiments/019-idle-links: links.c, many senders written from
serve/link.h's packet format alone (a second implementation), and the
quinn service and clients (quic/, built offline); run.py makes both
measurements.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
pure-serve --send OUTPUT=pure://host:port/INPUT sends an output's saved
sends to another node's --link-port with no handshake: the first packet
carries records. Keys come from the two nodes' keys (X25519, HKDF-SHA256,
AES-128-GCM), with a random run in each packet's header so no nonce
repeats across restarts; nothing is stored for them. What lasts is the
apps' own: the sender's outbox and the receiver's heard numbers
(pure_app_heard, new). The sender detects losses as RFC 9002 says, with
NewReno, and trims a send once the receiver says it stored it; the
receiver hands the owner whole records in order, which it applies from
the one after the source's last heard number, stopping at a gap, and
answers how far it stored with every acknowledgment. Stored bytes are
the bytes on the wire: an output and an input of one shape need no JSON.
server_links_udp kills the receiver, the sender, then both (-9) while
four clients buy: every saved order is applied once, with no handshake
(five runs, 10k-11k orders each). server_links_udp_lossy does it again in
a namespace of its own with the link's packets delayed, lost (3 %) and
reordered (10 %) by netem (three runs, 11k-12k orders).
|
| | |
|
| |
|
|
| |
and an SD card
|
| |
|
|
|
|
|
|
|
|
|
|
| |
091 measures four queries on pgbench's tables of 100,000 accounts, the
same rows and answers on Pure (registered queries, HTTP and JSON
included) and SQLite 3.46.1, through each step: items field by field in
leaves, folds and filters a leaf at a time, and one pass answering a fold
in a column for every row. Per-branch totals went from 398 ms to 4.9 ms
(SQLite 45, 14.3 with a DBA's indexes), a three-table join from 3.3 to
0.48 ms (1.47), and each rich account's rank from 9.2 s to 6.8 ms
(2.84 s, 7.6 ms with an index); counting accounts in credit takes 1.05 ms
where an index lets SQLite count 0.16 ms. Pure declares no index.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
benchmarks/bank/bank.aup keeps accounts, tellers and branches as keyed
lists and a history list; a transfer changes a row of each and appends a
history row in one event, as pgbench's TPC-B-like transaction does in
SQL. run.py makes a scale's rows and drives transfers through the load
generator's new transfer kind; sqlite-bank.c runs the same tables and
SQL on SQLite 3.46.1, a connection a thread, WAL.
With lists as trees, transfers run at 30.5k a second on 1,000 accounts
and 24.5k on 100,000 (RAM store, one connection thread, HTTP and JSON
included), where they ran at 3.2k on 30,000 before. Durable on the SD
card, Pure makes 1,339 a second and SQLite with synchronous=FULL 84 to
153. SQLite in process with no HTTP makes about 40k in RAM: recorded as
it is, with the write amplification of 100,000-row tables on the card
and ad hoc queries as what comes next.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
The roadmap's E2EE messenger made of the runtime's general parts, with
no messenger code in it: sealed, authenticated sessions between nodes
(086, 088), Vega on each node's display as a surface (design 010), the
app in Vega's examples. Vega's check runs two nodes through a recording
relay: a line typed on one display is drawn on the other, frames match
the vega command's, and the relay sees no text.
A line takes 0.79 ms at p50 and 1.1 ms at p99 from one node's program
to the other node's views with stores in RAM, both commits, the HTTP
delivery and the seal included (sealing adds nothing measurable); on the
Pi's SD card 3.5 ms at p50, the long tail being the card's syncs.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A sealed request cost three X25519 operations (about 0.5 ms), so sealed
sends drained at half the plain rate. A wire now keeps one HPKE context,
encapsulated at its first seal, and seals each request as the context's
next number (a failed seal or a restart starts a new session); the
header carries the session's encapsulated key and the number, so a
receiver opens a session once per connection thread (16 remembered, the
least used replaced) and each seal by its number, also after it restarts,
unasked. A sender with a key of its own seals in HPKE auth mode and names
that key in the header; a receiver started with --sender NAME=KEY takes
deliveries only from the senders it names, each sealed with its key
(403 sender otherwise). The body is now the ciphertext and its tag.
Sealed sends, sender authenticated, drain at 461-464k a second against
462-470k plain and 207-234k sealed a request at a time (Raspberry Pi 5).
The check adds: a wire's requests share one session numbered from 0; a
changed number or sender key does not open; a receiver restarted naming
its sender opens the same session again; another node claiming that
sender's name with its own key is refused and applies nothing.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
What the app model, snapshot reads and delivery had cost small reads
(126.7k -> 116.1k GETs a second) and idle memory (2.0 -> 4.5 MiB), how
the sampler found it, and the four steps and the crypto split that took
it back: 139.9k and 2.3 MiB. Measured again today beside axum 0.8.9 and
drogon 1.9.0 on the same Pi: a one-thread pure-serve answers small GETs
at 139k/s (2.2x axum and drogon a CPU second), 1 MiB GETs at 2,362/s
(axum 1,304, drogon 754), durable app writes at 21.3k/s (both about
200), with the least idle memory of the three.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
--seal OUTPUT=KEY seals each request's body whole to the receiving node's
public key with HPKE (RFC 9180, base mode: X25519, HKDF-SHA256, AES-128-GCM
where the CPU has AES, else ChaCha20-Poly1305), bound to the request's
source, first number and count; X-Pure-Seal names the suite. A receiver's
--key is a raw 32-byte file, made 0600 when missing, held in OpenSSL's
secure heap; it opens the body where it lies (400 seal when it does not
open) and zeroes the plain text when the request's ticket goes.
--sealed-only refuses unsealed deliveries (403 unsealed).
The check puts a recording relay between two apps: every order applies
once, no body shows an item or a field name, a flipped bit is refused, a
replayed request applies nothing twice, and a seal to another key does not
open. Its hosts choose their own ports (--port 0). Sealed sends drain at
233-235k a second against 469-470k plain (Raspberry Pi 5): three X25519
operations a request, 164 us each. The plain request path measured the
same as before.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commits grew the tree store's tail 256 KiB at a time, each growth a sync
that also committed the ext4 journal for the new size and blocks. A file
installed by compaction now carries written zeros up to where the next
compaction will come (4 x live + slack, plus a commit's margin), capped at
the size of the file it replaces, so a store never takes more space than
before and shrinks when its live data does. The cycle's commits then only
overwrite blocks the file has.
Store commits: 317.5 -> 385.5 a second, ext4 journal commits 109 -> 45 for
4,000 commits; durable app posts +5.3%, their p99 the same within this SD
card's noise. The new cycles test keeps sizes still between compactions
once the live data is whole, and shows the file shrinking with it.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Nothing a thread can give: with writes saturated the device's one line of
syncs is the bottleneck, so zeroing the tail or compacting on another
thread only queues that device work behind the commits, and the owner's
idle gap is shorter than one journal commit. Built without compaction the
store is no faster; without compaction and without tail growth durable
posts gain 1.4% and the p99 drops 24% (17.4 to 13.3 ms). New blocks are the
tail latency, so the way to it is a store that rewrites its space in place,
a ring cleaned forward instead of compacted into new files.
|
| |
|
|
|
|
|
|
|
|
|
| |
It would not, on this device. One file's syncs or four at once, the SD card
completes the same 510-690 commits per second, the latency growing with the
number of files. The scoreboard app with two owners, each with a store of
its own, took 23% fewer durable posts than one owner (median 16,230 against
20,974 a second) with 2-3x the p99: syncs queue, and each carries half the
events. One owner's group commit stays; cores serve reads, and writes grow
by making the sync itself cheaper.
|
| |
|
|
|
|
|
|
|
|
| |
The HTTP host's end-to-end checks (documents against the earlier server,
lending, kill -9 durability, JSON, timers, links, webhooks and answers) sat
in aup's tests, so a change to the host was verified from the toolchain.
They now live in tests/check.py; make check builds their apps with the
sibling aup and runs them after the unit tests, which stay beside their
parts. Design 009 states the rule for all three repositories.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The app model, the tree store and the HTTP host lived in
prototypes/connections among the earlier document runtime, the store was
known only by its format's name, and Vega compiled six Pure sources by path.
Now, as the README's principle has it:
core/ runtime, flow and the app model (module*.c), with what the core
asks of storage (state.h, store.h)
store/ the tree store (PURETRE3, PUREREC1)
serve/ pure-serve: HTTP host, outbound delivery, JSON in place
cli/ pure-run, pure-edit-module
A top-level make builds libpure.a (app model and store, for hosts to link)
and the hosts; make check runs the store's and the app model's tests. The
earlier runtime still builds from prototypes/connections against core/ and
store/ (check, check-tree and check-core pass). Design 009 records the rule:
parts by role behind libraries; git repositories split by coupling, so the
store stays here while its contract moves with the core, and Vega stays apart.
|
| |
|
|
|
|
|
|
|
| |
After each save the owner walked the whole outbox to find the sends
not yet queued, so a receiver long down made every save walk the
backlog. pure_app_outbox_newest visits saved sends newest first, and
the owner stops once every wired output reaches the number it queued
last, then queues what it found in order.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The final answer to each send of a wired output (a 2xx, or a 4xx but
408 and 429) becomes a message event of the given input: fields from
the answer's JSON object by name, other members skipped; a field the
answer lacks from the send's field of that name and type, so the
request's own id is there; a u32 http_status field is the status. It is
applied with the send's number from the wire's source, before the send
is trimmed and in the same save, so an answer that comes again after a
restart (the API repeating itself for the Idempotency-Key) applies once.
Sends that wait for answers go one per request; answers are saved with
the next round, or at once when none waits.
A server that writes an answer in pieces holds the second until the
first is acknowledged, and a delayed acknowledgement stalled every
answer 40 ms: the client now acknowledges at once (TCP_QUICKACK) after
each read. Against a payments API that answers 200 or 402, with the
sender killed (-9) midway, paid and declined orders matched the API's
answers one to one.
|
| |
|
|
|
|
|
|
|
|
| |
A wire to an https:// URL speaks TLS 1.2 or later through OpenSSL: the
receiver's certificate must chain to the system's CAs (or to the file
--send-ca names) and name its host (SNI and SSL_set1_host, or the IP
address). An untrusted receiver gets nothing: the send stays in the
outbox, the wire logs it and tries again every 10 s. pure-serve now
links libssl and libcrypto.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
pure-serve --send OUTPUT=http://host:port/path wires an output: after
every save the owner queues the new sends as JSON (fields and groups)
and a thread per output posts them in order with X-Pure-Source,
X-Pure-Sequence and Idempotency-Key. A 2xx, or a Pure receiver's
refusal, delivers; the owner then trims. 408, 429, 5xx and no answer
are sent again after a pause growing to 10 s, other answers every 10 s:
nothing is dropped. A Pure receiver applies a delivered message with
pure_app_deliver (now taking placed fields) and advertises
X-Pure-Batch: 256; the sender then posts up to that many numbers in a
row as one JSON array (X-Pure-Count), applied in order in one burst
with one save.
Delivery kept up with orders once batched (391 deliveries/s one at a
time; a backlog of 20,000 went at 60k/s). With four clients ordering,
killing the receiver and then the sender (-9) left the receiver's count
and total equal to what the sender saved, five times and under ASan.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
An output (input kind 5) is a message the app sends; SEND (72) takes
an output's field values in order in a handler, lays them out as a
message's payload and keeps it, numbered per output from 1, with the
event's changes: a failed event sent nothing, and a failed save takes
back every send since the last save with its numbers. A host sees only
saved sends (pure_app_outbox) and trims what its receiver has saved
(pure_app_trim, gone with the next save). pure_app_deliver applies a
message from a named source once: a number heard before succeeds
without applying, a refused message counts as heard, a failed save
forgets what it heard.
Sends, numbers and heard numbers are the app's links state, stored
whole in one runtime document (PLK1) with the states, in the same
commit. Outputs follow their names across module changes; a module
that drops or retypes an output with sends left is refused, and a
dropped output's number is kept for when it returns. Killing the
process at every commit stage leaves an event and its send, or a
delivery and its effect, together or neither. pure-serve answers a
post to an output or a timer with 404 no_input.
|
| |
|
|
|
|
|
| |
Measured with a build that grows its tail by 64 MiB once and never
compacts: durable posts +5-8% (64 connections), p99 about halved; the
slow syncs that remain come from the device.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
pure-serve no longer copies a JSON body's strings into the message's
payload. The ingress is the prefix, a table of each field's start and
size, and the body as it came; the connection thread scans the body
once, puts numbers into the prefix and records where each string is.
Only a string with escapes is unescaped, over itself (it never grows).
A group of fields comes as an object inside the body and, for a fixed
field, as its path in the query (?to.zip=7). A group where a field
belongs, a value where a group does, an array or true/false/null is
json_type; an unknown member, inside a group too, json_field.
The string scan and UTF-8 checks now pass over eight bytes at a time
while none needs a look (pure_utf8_step, so every text check gains).
On the Pi, posting {"name":"k","text":"x..."} to a stateless app went
from 652-655 to 3,737-3,953 requests/s with 256 KiB of text (server
CPU 1.52 to 0.25 ms a request, peak RSS 5.2 to 3.2-3.4 MiB) and from
29.1k to 76.3-76.7k/s with 4 KiB; the load generator gains a json
method and benchmarks/http/echo.aup is the app.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A tree store commit wrote its bytes over the written zeros of the tail
and its head into a slot at the start of the file: two ranges, and on
the Pi's SD card a sync of two ranges costs about twice one (2.4 ms
against 1.2 ms). A commit now writes a 128-byte record (its head) at
the published end and its bytes right after it, then syncs once. Every
16 commits the head also goes into the older slot. The newest commit
is found from the newer slot's head, or from the newest head this
process has seen in the file while its record is still there, along
the records whose bytes are whole; a torn record or bytes end the walk
at the commit before, which the next writer overwrites. Slots of this
format are PURETRE3; files of the earlier one are read and continued.
Store commits went from 220-233/s to 371-396/s (p50 2.5 to 1.3 ms),
and the scoreboard's durable posts from 13-16k/s to 21-26k/s with 64
connections and from 2.2-2.4k/s to 2.9-3.9k/s with 8, measured
alternately with the previous build. The torn head check now tears a
slot (the walk goes on from the other) and the newest record (a new
process sees the commit before); a new check walks 42 commits from the
slots, in a fresh process too, and stops at a damaged record.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The owner publishes a snapshot of the saved states after every save,
before it hands back the answers; connection threads take the newest
one each round and show views from it with a reader app of their own.
A read made after a write's answer sees the write, nothing unsaved or
undone is ever shown, and reads no longer wait while the owner applies
events or syncs. Apps with states always get an owner thread, even
with one connection thread. A round's inputs go to the owner first;
its views are shown and then answered together (answering each one as
it was shown made one thread's 20-row page 8% slower).
On the Pi 5, with 8 connections posting durable scores and 56 reading
the total, reads went from 10-16k/s at p99 13-29 ms to 107-149k/s at
p99 0.6-0.7 ms, and posts stayed in range; axum and drogon, which sync
under their lock, read 1.1-1.3k/s at p99 111-156 ms. A snapshot of a
changed 1,024-item list costs 1.2 us (65,536 items: 87 us) against a
3-4 ms save.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
A view that shows a text state as it is (pure_module_text_state) now goes
out as the state's stored extents, as a document read does: the answer
holds the store mapping, charged to the budget, until it is sent and its
ticket released; a state with edits not yet stored is copied as before.
Documents and views share one loan path, and the extent array grows as
needed. A 1 MiB note read by 8 connections: 776-817 answers a second
copied, 1,875-1,901 lent, p50 9.2 to 3.5 ms, peak RSS 10.3 to 5.7 MiB.
The load generator learns GET /views/page.
|
| |
|
|
|
|
|
|
|
|
| |
A sent message belongs to its event's transaction in a runtime-owned
outbox; hosts deliver committed messages in order, and a receiver keeps
the last sequence per source in the same commit, so the effect is exactly
once. The note records the design and its order of work; it waits behind
the roadmap's short-term items and goes with outbound calls, which stand
on the same outbox.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
The scoreboard, 22 lines of aup served by its views and inputs, against
the same service written by hand in axum and drogon on the Pi: durable
posts about 76 times as many, reads level with drogon and ahead of axum
with the least CPU and memory. The note keeps what the profile showed on
the way (the kernel's share of a single thread, integers through printf,
an owner that saved too early) and what comes next. run.py gains
pure-serve-2 and waits for a server that prints its port before it
listens.
|
| |
|
|
|
|
|
|
|
|
|
| |
The zero tail wrote as many bytes as the commits under 64 writers, yet
turning it off changed nothing beyond the SD card's own noise and made
8 writers wait longer, so it stays until faster storage is measured.
Overlapping a sync with the next burst would win a few percent at most.
The second sync of a new process's first commit is rare with one owning
server. The larger lever is write amplification: a 16-byte edit rewrites
its whole leaf and catalog path.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A drogon 1.9 service with the axum baseline's contract joins the benchmark
(memory, fdatasync and crash-safe modes, three IO threads), built from
Debian's packages unpacked under .deps without installing them. drogon
listens once per event loop, so the service picks a free port itself:
with port 0 every loop got a port of its own and the load reached one.
The load generator has one core and saturates near 150k answers a second,
so the harness now also reports answers per second of server CPU time
(all threads). Over three rotated rounds: small GETs 151.6k/s on three
shards (drogon 145.3k, axum 120.9k) and 128k per CPU-second on one thread
(drogon 65k, axum 62k); 1 MiB GETs 2,794/s (axum 1,372, drogon 708);
crash-safe PUTs 5.1-5.8 times axum and 13-30 times drogon; peak RSS
5.0 MiB against 14.4 and 76.5 MiB. --workloads is now a regular expression.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A commit used to sync its nodes, then write the head and sync again: the
order of writes was forced by syncs. Now the head carries where the
commit's bytes start and their digest, and one fdatasync makes the bytes
and the head durable together. A crash during it can leave a head whose
bytes did not all reach the device; its digest then fails and the older
slot stands, which is safe because a writer only builds on a durable head.
Commits overwrite written zeros kept after the published end, so the sync
changes neither the size nor the block map and needs no journal commit.
On the SD card this is 2.4 ms per commit against 9.1 ms for two syncs;
the store commits 289-304 times a second instead of 94-109, and pure-serve
answers 2,012-2,611 durable PUTs a second against 583 for crash-safe axum.
A process remembers the heads it published or checked, so it reads a
commit's bytes at most once and never its own. Its readers pass over the
head it is still making durable and so never answer with a commit a crash
could take back. Head slots get a block each and the format becomes
PURETRE2; a handle's refresh reads the slots through its mapping.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
A read of a stored document now answers with the store mapping's extents
through writev, without copying the document. pure_draft_mapping takes a
reference to the mapping a draft opened through a handle reads from; the
answer keeps it until the last byte is sent, so the version read goes out
intact even if the document changes or a compaction replaces the file
meanwhile. Replies the mapping does not hold are copied as before.
pure-serve's output is now a list of extents (the head first). One thread
serves 2,761 1 MiB reads per second (axum: 1,380) and 128k small ones.
|
