russell_sears

4 ms·

1.
by russell_sears 2y ago
My grandfather used to work on telephone field network boxes (e.g., the big boxes full of messes of wires in residential areas). One job was to look for outliers in the network, and they spent time studying areas with an unusually large num
2.
by russell_sears 3y ago
This looks like the sort of bug I'd write back when I used mutexes to write I/O routines. These days, I'd use a lock-free state machine to encode something like this: NOT_IN_CACHE -> READING -> IN_CACHE (the re
3.
by russell_sears 8y ago
You might also be interested in JOL/Overlog, which was the predecessor to Bloom: https://github.com/tcondie/jol We built quite a few prototypes on top of it. The language implementation is a research prototype, b
4.
by russell_sears 8y ago
You can guarantee termination by limiting bit precision, but it is more interesting to limit the set of operators your program uses instead. For instance, you might have an operator that translates from hostname to IPv4 address. That functi
5.
by russell_sears 10y ago
We really need a global carbon credit system (with stiff tariffs for countries that do not participate). Cut net greenhouse emissions by a few percent a year until net emissions are negative and atmospheric CO2 and methane are back to pre-
6.
by russell_sears 10y ago
I still don't understand how you come to the conclusion that the attack won't work. The bolded text in the blog post reads "the middle relay will pick a different exit relay", and should read "the client will pick a
7.
by russell_sears 10y ago
I read this explanation a few times, and am afraid I still don't understand why Tor is not vulnerable to this attack. Let's say I'm an attacker that provides enough bandwidth to compromise N% of client circuts, and I can dis
8.
by russell_sears 10y ago
There are two effects at play. If you're reading a multiple of a flash page at a time, then you can get random bandwidth to match sequential. Under the hood, decent SSDs actually round robin sequential reads across all of the underly
9.
by russell_sears 10y ago
Hi. First author of the bLSM (Yahoo) paper here. You've pretty much nailed it: The trick is getting latency and volume at the same time. If you're willing to wait an hour to look at event logs, then Hadoop is good enough, but w