nivertech

Joined 16 years ago

Karma

1,978

Submissions

1,840

Member since

Apr 23

[ my public key: https://keybase.io/nivertech; my proof: https://keybase.io/nivertech/sigs/TSPizHrahhxd9Q1ZrIOFu3_73baE_0_XgUQTWapr4Zs ]

  - Distributed Systems
  - DDD / CQRS / Event Sourcing
  - Erlang/OTP, Elixir/Phoenix/Absinthe/LiveView, Elm
  - kdb+/q
  - GraphQL, MQTT, Pub/Sub, WebSockets, Long Polling, etc.
  - PostgreSQL, Redis, DynamoDB, Riak
  - AWS / GCP
  - GPU Computing (OpenCL, CUDA)
  - Agent-based Modeling/Simulations (ABM), NetLogo

View on Hacker News

Recent submissions

Comment · 19 days ago

Compiler compiled it.

Comment · 21 days ago

> Indo-European ethnicities/language families

language families, not ethnicities

Comment · 1 month ago

yes, that was another one I had in mind.

  /ən ˌɛs.kjuːˈɛl ˈkwɪri/
  vs
  /ə ˈsiːkwəl ˈkwɪri/

Comment · 1 month ago

  “a San Francisco startup” /ə ˌsæn frənˈsɪskoʊ ˈstɑɹtˌʌp/
  vs
  “an SF startup” /ən ˌɛsˈɛf ˈstɑɹtˌʌp/

i.e. you have to mentally pronounce the phrase before choosing which form of the indefinite article to write

Comment · 1 month ago

> Q. How do you port an OS from one language to a new one, pre-LLM?

A. you reduce impedance mismatch, but making the new language using similar concepts and mechanisms

You're right, this table is just a coincidence, probably derived from the standard math notation, and used in many PLs like FORTRAN and Python ;)

  PDP-11        C
  
  INC R         ++i
  DEC R         --i
  ADD src, dst  dst += src
  SUB src, dst  dst -= src

  (R)+          *p++
  -(R)          *--p
  X(R)          p[x]
  @(R)+         **pp++
  @X(R)         *p[x]

  BR label      goto label  ; near jump
  JMP label     goto label  ; far jump

Comment · 1 month ago

It's clear that the authors were either inspired by the PDP-11 ISA and/or designed the language to make porting UNIX from PDP-11 Assm to it easier.

UNIX for PDP-7 in Assm -> UNIX for PDP-11 in Assm -> UNIX for PDP-11 in C

https://news.ycombinator.com/item?id=42644851

and

https://news.ycombinator.com/item?id=43245166

Comment · 2 months ago

Now we know who built the Silo

Comment · 2 months ago

> 50 is the hardware limit

this can't be true, If that hardware can run Django and Phoenix, it for sure can handle orders of magnitudes more concurrent clients than 50 even before you start tuning the OS and the runtime.

> Playwright headless Chromium

maybe you mean you're unable to generate more than 50 concurrent clients using Playwright+Chromium - but that's the wrong tool for loadtesting.

Loadtest should run from a different machine or cluster of machines/VMs. You can generate up to ~ 64K outgoing connections per IP address / alias. With 17 IP addresses/aliases you can generate up to C1M.

Comment · 2 months ago

Your "Concurrency" benchmark ends at 50 concurrent clients. Continue testing at C1K, C10K, C100K, C1M, C2M, etc.

That's for a single node.

Then try multiple nodes / cluster - C1M, C2M, C10M, etc.

Also, all concurrent clients should send periodic pings / keepalives and do something useful (e.g. some CPU work, not just being idle).

Then you'll have to change your conclusion from:

> The data answers my original question: yes, Django LiveView holds up against Phoenix LiveView. In day-to-day operations they are practically equal.

> The data answers my original question: yes, under *toy* load Django LiveView holds up against Phoenix LiveView. For *toy* operations they are practically equal.

Comment · 2 months ago

It pretends to be serious with 5 tables / charts:

Django LiveView vs Phoenix LiveView: a real benchmark

https://en.andros.dev/blog/80134668/django-liveview-vs-phoen...

Comment · 2 months ago

TL;DR: this is a toy example, so it doesn't require BEAM VM

Hard to take this seriously without the loadtest with the the large number of concurrent users.

For a number of users concurrently reading a long-tail blog post - you will get good results even using DotCom era HTTP polling.

---

See:

Django LiveView vs Phoenix LiveView: a real benchmark

https://en.andros.dev/blog/80134668/django-liveview-vs-phoen...

Comment · 3 months ago

It doesn't matter CRUD, REST, 2PC, or middleware in general. From my PoV they're all technical solutions used to avoid deep modeling of the domain and its workflows, and instead trying to model the real world using generic primitives without understanding it first, or maybe with surface-level understanding only

The dichotomy is generic vs custom/bespoke:

- generic (CRUD / REST / RDBMS+SQL+ACID Tx / Tx Scripts / 2PC / BPMN / Admin UIs) - also SmartUIs/Fat Clients (VB, Delphi, SPAs, etc.)

- custom (DDD / CQRS/ES / Sagas / Process Managers / Durable Executions / TBUI)

Also, learning data engineering and BI helps one better model the real world than generic methods. It's somewhat similar to CQRS/ES, but with much more prior art