A 1970s Cray-1 hard drive has been imaged
blog.archive.org
A 1970s Cray-1 hard drive has been imaged
1–10 of 38 posts
Re: A 1970s Cray-1 hard drive has been imaged
#2Gosh, some actual "hacking" in Hacker News.
Just waiting for the comments asking "what does this have to do with startups" ...
Re: A 1970s Cray-1 hard drive has been imaged
#3Norton Ghost?..
Re: A 1970s Cray-1 hard drive has been imaged
#4Gosh, some actual "hacking" in Hacker News. Just waiting for the comments asking "what does this have to do with startups" ...
I'd like to see more hacking news and less stuff about startups as well.
Great article, it's incredible no more Cray-1 software has been preserved.
Re: A 1970s Cray-1 hard drive has been imaged
#5Specs on storage module drive: http://ia600707.us.archive.org/32/items/9760_9762_oct73/9760... .
The unit has an average seek time of 30ms in 1973. Today, in common 7200 RPM drives, it seems to be around 8-9ms. There is only a factor of improvement of 3 to 4 times.
I find this pleasantly surprising.
Re: A 1970s Cray-1 hard drive has been imaged
#6That's pretty impressive. Not only did they image the disk, it was a disk that had had a head crash and one damaged head which they managed to restore to functioning well enough to capture the data.
Now of course the big question: what is on that disk?
Re: A 1970s Cray-1 hard drive has been imaged
#7Specs on storage module drive: http://ia600707.us.archive.org/32/items/9760_9762_oct73/9760... . The unit has an average seek time of 30ms in 1973. Today, in common 7200 RPM drives, it seems to be around 8-9ms. There is only a factor of improvement of 3 to 4 times. I find this pleasantly surprising.
Why pleasantly surprising?
Re: A 1970s Cray-1 hard drive has been imaged
#8Specs on storage module drive: http://ia600707.us.archive.org/32/items/9760_9762_oct73/9760... . The unit has an average seek time of 30ms in 1973. Today, in common 7200 RPM drives, it seems to be around 8-9ms. There is only a factor of improvement of 3 to 4 times. I find this pleasantly surprising.
Latency has been scaling worse than almost any other factor for nearly the entire history of computing. The latency/throughput ratio for storage, networking, memory, cache and almost everything else has been rising continuously since the 70s and is likely to carry on rising.
This is going to have big consequences for the way that we design systems in the future. Transferring a large amount of data is going to be cheap and fast. Seeking, handshaking, back-and-forth and any other latency sensitive operations are going to be slow and expensive. This has already played a big role in algorithm design in HPC, and is going to start being felt to a much larger degree in the wider field over the next decade. As the latency/throughput ratio gets bigger, the tradeoffs behind optimal system design will change.
30ms for a Cray 1 was only 2400 clock cycles. 8ms for a modern CPU is around 30 million. That's a big change.
Re: A 1970s Cray-1 hard drive has been imaged
#9Didn't I read about someone who built a replica of a Cray, but struggled to find actual software for it? This could be exactly what they needed. I think the story of the Cray replica (scaled model) was featured on HN.
Re: A 1970s Cray-1 hard drive has been imaged
#10Didn't I read about someone who built a replica of a Cray, but struggled to find actual software for it? This could be exactly what they needed. I think the story of the Cray replica (scaled model) was featured on HN.