Student, programmer, sysadmin, entrepreneur, lisper.
Definitely opinionated, I encountered a surprising amount of different systems in my life so far, leading to rather eclectic set of skills. My current favourites are Common Lisp and Erlang.
Trying to finish some degree (officially HND MSci from University of Aberdeen, but who knows how it will end...) and to get to Google ;-)
|
May
16 |
|
comment |
performance - multithreaded or multiprocess applications The biggest possible differences would be in scheduling, and that can be modified with cpu affinity of the task, pinning each task (thread/process) to specific cpu. More differences could be from actual work done by server process and how it interacts with network part. There can be differences related to how memory is accessed, but less related to thread/process difference as to sharing memory in general on a NUMA system - though it's possible that the current kernels have support to make copies of code segments in different NUMA zones (they did have that for kernel itself in the past). |
|
May
16 |
|
answered | Linux: forward incoming connections from subdomain to port? |
|
May
16 |
|
answered | performance - multithreaded or multiprocess applications |
|
Apr
13 |
|
awarded | Yearling |
|
Apr
13 |
|
awarded | Yearling |
|
Apr
12 |
|
awarded | Necromancer |
|
Feb
27 |
|
awarded | Caucus |
|
Feb
14 |
|
awarded | Critic |
|
Feb
14 |
|
comment |
Why doesn't TrueCrypt support system encryption on non-Windows machines? They don't have to figure ext2/3/4/whatever - proper FDE works at block level, and uses a block layer driver to decrypt. I suspect windows has this done through the boot storage driver and later registering truecrypt driver as filter between disk and filesystem drivers. |
|
Feb
6 |
|
comment |
Production-ready, highly reliable filesystems on Linux: ext4 ext3 XFS or JFS (or ZFS)? Since then, ZFSonLinux became a pretty good solution, and in fact I'm running it in production on my day job workstation. It works pretty good, but you need to be careful when setting it up. |
|
Jan
9 |
|
comment |
Why was Tanenbaum wrong in the Tanenbaum-Torvalds debates? @mouviciel The only cases of L4 in Android are inside radio firmware on Qualcomm chips, which runs on separate cpu. The Mach in OSX/iOS is 2.5 which suffers from all maladies of microkernels and nearly none of benefits (and has a lot of monolithic stuff in kernelspace to get any performance). Spring OS, from Sun, did get microkernel right, similarly QNX, possibly BeOS, and few others. |
|
Jan
9 |
|
comment |
Why is Linux called a monolithic kernel? Windows NT was always a hybrid system. The kernel might be not exactly hybrid, but you run into question of what you count as part of it (winapi, for example, is implemented as user-space service) |
|
Jan
9 |
|
comment |
Why is Linux called a monolithic kernel? Actually Linux these days is a hybrid kernel, although sharing (kinda) single memory space - IIRC during 2.5.x, the kernel switched from much more classic "monolith" into a bunch of async/event-triggered processes running in kernel space (something that was common in Mach2.5-derived OSes, and became official part of Mach in 4.0) |
|
Nov
8 |
|
awarded | Citizen Patrol |
|
Oct
23 |
|
comment |
Why didn't "cloud computing" offer appear earlier? The name "cloud" basically comes from the ancient use of "cloud" on network diagrams...and yes, I remember such clouds even on some IBM mainframe-related ones ;) |
|
Oct
7 |
|
comment |
Greenspun's Tenth Rule, does every large project include a Lisp interpreter? Actually, they stopped using GOAL because they got bought by company whose codebase was in C++. Also, GOAL was quite the lisp. Don't assume lowest-denominator online tutorials and university lectures are correct :) |
|
Oct
7 |
|
comment |
Greenspun's Tenth Rule, does every large project include a Lisp interpreter? Seems like relevant only to the most ancient lisp environments of late 50s. Personally, I found Common Lisp's functions for dealing with bits probably one of the best (with main competition being Erlang). Arrays, hashtables, structs are all common. |
|
Sep
22 |
|
awarded | Yearling |
|
Jul
24 |
|
comment |
Why aren't there other programming languages that compile to Python bytecode? @Rig It's because Python's OOP was very, very barebones, so it exposed the self argument which is "hidden" in Java, C++, Ruby and others. So you have parts of the internals showing up. |
|
Jul
24 |
|
answered | How to compile a "tun.ko" file for OpenVPN? |
