OpenTSDB: http://opentsdb.net/
StatsD: https://github.com/etsy/statsd/ (description here - http://codeascraft.etsy.com/2011/02/15/measure-anything-meas...)
OpenTSDB and StatsD seemed great for getting TS data in and producing nice dashboards, but they didn't seem to fit our needs for performing custom analytics on the data.
At the moment, we're leaning towards leveraging Cassandra based on our scalability requirements. Check out http://rubyscale.com/blog/2011/03/06/basic-time-series-with-... and http://www.datastax.com/dev/blog/advanced-time-series-with-c... to get an idea on how cassandra can help.
Our performance issues with mongo largely stemmed from our poor use of indexes - we defined a lot of indexes because how we needed to query was a very organic and undefined process as we got new analysis requirements. Because we would have to frequently go back and compute new feature vectors across the whole (or large parts of) the dataset, we weren't able to implement a lot of the aggregation capabilities you'll see implemented in many other time series schemas.