Hi Luke, On Sat, Jun 13, 2009 at 2:25 AM, Luke<[email protected]> wrote: > > Congrats Mateusz and crew!
Thanks a lot ! > This service looks great. The authentication interface looks > reasonable as well. One optimization can be made is that the scanner > and mutator interface doesn't need authtoken besides the open_* > interface, since scanner id and mutator id are already random 64-bit > integers. It'll save you an additional authtoken lookup at the thrift > broker side. > > Do you mind that I merge in the API accordingly (the authtoken is > ignored by default.)? Comments/suggestions from everyone are welcome. Sure. I'll be making the set of patches public starting probably tomorrow (it's 2:57 AM already here and I'm going to sleep in 5 minutes). WRT the authentication API, I was also wondering if the code would benefit from making Thrift session states persistent with Berkeley DB, so I'm very keen to know your take on this? Right now the approach is to keep everything in memory which somehow does not go well along with my personal taste. > One question, do you charge traffic even within ec2? If you make > intra-ec2 traffic free, it'll be a lot more attractive. The pricing is going through the review process and the feedback so far means it has to change. Hopefully, the prices will get updated over a weekend. With regard to intra EC2 traffic, this is a very good point indeed. I think this might be one of important factors and yet it somehow escaped our creative process ;-) Thanks for noticing it. best, Mateusz > > __Luke > > On Fri, Jun 12, 2009 at 3:15 PM, Mateusz Berezecki<[email protected]> wrote: >> >> Dear hypertable developers and users, >> >> Sometimes installing and maintaining Hypertable can be tough thing to do, >> and we are aware that many of you, the users, are not accustomed to working >> with this new approach to handling large volumes of data. >> >> Given these difficulties, we have decided to work on helping you >> with some of the issues. >> >> As a first step, we have just launched an alpha version of our new >> service http://www.episcale.com - which is essentially a hosted >> hypertable cluster. When you register you get an IP and port and you >> can use Hypertable as if you would have installed it on your own. >> >> But we have done a bit more to make it easier: >> >> - you only pay per usage (storage and queries) >> - you can connect from many programming languages such as >> C++, Perl, Ruby, Python, Java, PHP, Erlang, C >> - we will monitor performance for you so no need to worry >> about what is going on >> >> We would like to work hand-in-hand with you to make your application >> run with episcale/hypertable - if you run into issue please email us >> anytime at [email protected] . >> >> best regards, >> Episcale Team >> >> > >> > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Hypertable Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/hypertable-dev?hl=en -~----------~----~----~----~------~----~------~--~---
