> Does anyone think it's odd that an Acer laptop could write these files to disk faster than JSTOR could serve them?
Why would that be odd? SATA = 3 Gbps throughput with minimal overhead, Ethernet = 1 Gbps with lots of overhead (IP headers, Ethernet headers, HTTP headers)
It sounds like JSTOR's servers aren't really optimized for high article download rates, to the point that his one laptop accounted for a significant part of normal continental US load. They probably had some bottleneck in the system that they never noticed before — maybe their logging infrastructure was absurdly slow or something.
Did it specify that he was writing to local disk? If I did a stunt like this, I'd probably try to write to some cloud storage like Amazon S3. I'd bet that Amazon's servers can drown JSTOR's, particularly if there's a big pipe like MIT's in-between.
Why would that be odd? SATA = 3 Gbps throughput with minimal overhead, Ethernet = 1 Gbps with lots of overhead (IP headers, Ethernet headers, HTTP headers)