21,218 reputation
12347
bio website ripple.com
location Oakland, CA
age 44
visits member for 3 years
seen 1 hour ago

I am Chief Cryptographer at Ripple Labs and one of the architects of the Ripple payment system.


Mar
8
comment Different results from GnuPG and OpenSSL when using AES256 encryption
Both of these tools internally use AES256, but their output formats are completely different. Can you be more specific about what you're trying to do? Do you really need deterministic encryption where the same input always produces the same output?
Mar
7
comment using a gre tunnel as default route
You cannot allow a tunnel to become the route to its own destination or the tunnel implodes.
Mar
7
comment Why does RAID5 with an odd number of data disks have poor write performance
let us continue this discussion in chat
Mar
7
comment Why does RAID5 with an odd number of data disks have poor write performance
@psusi: You're taking us in circles. Right, we all agree on that. Again, the question is: After this aggregation, do you think more writes will be an even multiple of 2MB or more writes will be an even multiple of 2.5MB?
Mar
6
comment Why does RAID5 with an odd number of data disks have poor write performance
@psusi: I don't understand why you keep taking us in circles. Yes, there's aggregation. But after aggregation, if the write is an even multiple of the stripe size, no read before write is needed. Otherwise, a read is needed before the write. Do you think more writes will be even multiples of 2MB or will be even multiple of 2.5MB? If more writes are even multiples of 2MB than 2.5MB, then a RAID with a stripe size of 2MB will perform better since less data will need to be read in order to write.
Mar
6
answered Do CDMA based time servers passively read timing information from the cell network?
Mar
6
comment Fast, but extremely spikey network transfers
I don't see any evidence that it's not pulling the data off the drive constantly. Do you have any reason to suspect that during the times when the network traffic is lower, the disk I/O throughput is also lower? (Imagine if you're receiving pumpkins one at a time and sending them out in trucks of 1,000. You'll see a pause, then a truck of 1,000, then a pause, then a truck of 1,000. But pumpkins are arriving at an even rate. More trucks wouldn't make pumpkins arrive faster nor does pauses between trucks indicate pauses in arrival.)
Mar
6
comment Why does RAID5 with an odd number of data disks have poor write performance
@psusi: The reason would be that fewer reads are required to write because fewer partial stripe writes are necessary. Fewer partial stripe writes are necessary because after aggregation, at the RAID layer, more writes will be even multiples of 2MB than of 2.5MB.
Mar
6
comment Why does RAID5 with an odd number of data disks have poor write performance
@psusi: That's nonsense. Whether an operation is sent to the drive as a pure write or a "read-then-write" depends on whether it can be aggregated to a multiple of the stripe size. So that aggregation has to be done before drive level, at the RAID level.
Mar
6
comment Linux: find total memory in use…without free(1) or /proc
By memory, you mean physical memory, right? But what precisely do you mean by "in use"?
Mar
5
answered 2 BGP Upstreams, One sending default route, another Full Table
Mar
5
comment Is it possible to generate RSA key without pass phrase?
Your command line tells openssl to encrypt an existing key. That sounds like something other than what you want.
Mar
5
comment server not responding
It depends on exactly what his problem is. If his problem is that he has lots of useless data suck in memory, swap would solve it. However, if the amount of useless stuff is unbounded and increasing, then it would just delay it.
Mar
5
answered server not responding
Mar
5
comment server not responding
It's not surprising that kswapd is having a hard time making free memory with no swap! Making some swap would make kswapd's job much easier. You need at least 2GB of swap or so to get any kind of sanity on a large system.
Mar
5
answered How to interpret Cisco IOS web “signal to noise(dBm)”?
Mar
5
comment Why does RAID5 with an odd number of data disks have poor write performance
Right, and when that final scheduling is done, do you think there are more operations that align on multiples of 2MB or operations that align (start or end) on multiples of 2.5MB? That is, which underlying stripe size do you think will require the greatest number of reads-before-writes? Stripe sizes that are powers of two better adapt to buffers that are powers of two and most software uses buffer sizes that are powers of two.
Mar
3
comment Why does RAID5 with an odd number of data disks have poor write performance
After they split and join operations, do you think they wind up with more 2MB operations or 2.5MB operations?
Mar
3
comment Why does RAID5 with an odd number of data disks have poor write performance
So you're saying that bulk writes with sizes that are even multiples of 2.5MB are just as common as bulk writes that are multiples of 2MB? That seems a bit implausible to me.
Mar
3
comment Why does RAID5 with an odd number of data disks have poor write performance
@psusi: Whether it is 2MB or 2.5MB makes a reasonably significant difference. Requests from higher levels tend to be for aligned blocks. With 2.5MB blocks having to go out to the RAID, a higher fraction of writes won't be complete blocks and more writes will require a preceding read. Let me put it to you this way -- if we did a survey of write sizes from major applications, do you think more write clusters (sequential writes from the same source) would be multiples of 2MB in size or multiples of 2.5MB in size? Even if you use, say, a 1MB buffer, two buffers will fit perfectly.