ants_a

5 ms·

1.
by ants_a 7mo ago
This is not a big database usecase. It just needs one to not do silly things like opening a new database session for every query when it's well documented that this is expensive.
2.
by ants_a 7mo ago
pgbouncer added support for prepared statements a couple years back.
3.
by ants_a 7mo ago
Over here we use a PKI cert for that. A smartcard providing the root of that trust is provided by the government after verifying your identity using the typical stuff used for identity documents (any biometric data on file, birth certificat
4.
by ants_a 8mo ago
What you are describing here does not match how postgres works. A read on the replica does not generate temporary tables, nor can anything on the replica create locks on the primary. The only two things a replica can do is hold back transca
5.
by ants_a 8mo ago
There is no backpressure from replication and streaming replication is asynchronous by default. Replicas can ask the primary to hold back garbage collection (off by default), which will eventually cause a slow down, but not blocking. Laggin
6.
by ants_a 8mo ago
These are limitations in the current PostgreSQL implementation. It's quite possible to have consistent commits and snapshots across sharded databases. Hopefully some day in PostgreSQL too.
7.
by ants_a 8mo ago
Postgres setups are typically based on physical replication, which is not an option on MySQL. My testing shows the limit to be about 177k tps with each transaction consisting of 3 updates and 1 insert.
8.
by ants_a 10mo ago
For updating a single resource where the order of updates matters the best throughput one can hope for is the inverse of locking duration. Typical postgres using applications follow the pattern where a transaction involves multiple round tr
9.
by ants_a 11mo ago
I'm wondering if it would make sense to integrate the rim, motor and wheel bearing into a single assembly to save weight and cost. That combined with the weight and packaging benefits of not having half shafts and differentials might m
10.
by ants_a 1y ago
In that snippet are links to Postgres docs and two blog posts, one being the blog post under discussion. None of those contain the information needed to make the presented claims about throughput. To make those claims it's necessary to
11.
by ants_a 1y ago
Triggers are not even particularly slow. They just hide the extra work that is being done and thus sometimes come back to bite programmers by adding a ton of work to statements that look like they should be quick.
12.
by ants_a 1y ago
So my display aspect ratio is 2.5dB. Or is it 5dB because it's not measuring power?
13.
by ants_a 1y ago
https://www.youtube.com/watch?v=vz-dhwSpjOw
14.
by ants_a 1y ago
That thread is indeed about the same issue. I don't think anyone has done a more concise writeup on it. Core of the issue is that on the primary, commit inserts a WAL record, waits for durability, local and/or replicated, and then
15.
by ants_a 1y ago
Interesting why this magic would be needed. Vanilla Postgres does support quorum commit which can do this. You can also set up the equivalent multi-AZ cluster with Patroni, and (modulo bugs) it does the necessary coordination to make sure t
16.
by ants_a 1y ago
Yes and no. On the primary durability order and visibility order are different. So an async transaction that starts committing later can become visible to readers before a sync transaction that comes before it. A read-write sync transaction
17.
by ants_a 1y ago
There is a paper exploring this concept: https://cs.uwaterloo.ca/~kdaudjee/ED.pdf UI wise it does not make sense to have this distinction, as the window to get durability is a small fraction of a second. But for concur
18.
by ants_a 2y ago
Consumers already have the choice of not buying the most powerful card. Increasing die size to run cores in a more power efficient regime is not going to work, because a) the chips are already as big as can be made, and b) competition will
19.
by ants_a 2y ago
LLMs are not a great source for this type of information. This mess in no way articulated in what way savepoints and transactions are different. Savepoints are fully equivalent to nested transactions with the constraint that only one is con
20.
by ants_a 2y ago
Not disagreeing, but for more context. These blog posts also have to be taken with a grain of salt as the authors obviously have their own project to promote, coloring their point of view. The algorithms detailed in the graph DB post can be
21.
by ants_a 2y ago
Javas problem is a culture of overly generic solutions (vendor neutral!) that then try to fix architectural performance issues by liberally adding caching. This makes both the startup time and consistency way worse than it needs to be.
22.
by ants_a 2y ago
No, and it feels unlikely that they will either.
23.
by ants_a 2y ago
For RDBMS we typically consider millions to be tiny to small. A billion is somewhere on the boundary of medium to big.
24.
by ants_a 2y ago
Traffic on I-10 at Arizona state line is 31k vehicles per day, with peak hour being 4.3k vehicles per hour. Assuming 350Wh/km, 120kW average charging rate and 90% utilization 3.2 charger stalls per mile per 1k vehicles/hour are ne
25.
by ants_a 2y ago
SELECT date_bin('5 minutes', log_time, '2000-01-01') log_time, AVG(value) avg_value FROM logs GROUP BY 1
26.
by ants_a 2y ago
A comparable product from Cerebras is rumored to cost around $2-3M.
27.
by ants_a 2y ago
That's not the motherboards fault, the PCIe controller is on the CPU and needs to have corresponding I/O pinout on the socket. And even from the CPU manufacturer side, it's not artificial market segmentation. The I/O die
28.
by ants_a 3y ago
For well scaling workloads optimizing for efficiency is optimizing for performance. For serial workloads even one P-core is enough. There aren't too many workloads in the middle that scale but only to a few cores.
29.
by ants_a 3y ago
Networking doesn't have to have high latency. You can buy network hardware that is able to provide sub-microsecond latency. Physical distance still matters, but 10% of typical NVMe latency gets you through a kilometer of fiber.
30.
by ants_a 3y ago
Nice, I hacked together something similar many moons ago, it looks like some fine folks have kept on maintaining and improving it: https://github.com/fschulze/sqlalchemy_schemadisplay Useful to have in a CICD pipeline

More ›