I'm in the industry, so I have a little more background on this. They spent about 40M GBP building the system, and it's only been used for two years. It was (entirely?) outsourced to Accenture. Other reasons why the system sucks: It can only handle about 10,000 orders/second, and has latency numbers that are incredibly high (5 milliseconds+).

Looking at other exchanges, there are trading platforms that have been able to last 10+ years while scaling quite well.

TradElect was/is a project management and technical disaster.

So Google used horizontal partitioning to split load across servers? Wow, that's rocket science. None of us in the database community have thought of doing this before. :-) But, if you want to find some news here, you can. One nice thing that Google did recently was to donate their horizontal partitioning code for Hibernate to the open source community. Hibernate Shards definitely needs a lot of work to get it to the point where it does a lot of stuff that people would want, but, hey, release early and often!