Thoughts on current functionality, requests for future functionality
Here are some thoughts on current functionality as well as some questions on what will be offered in the future:
1) There are many references to "lost images" and non-persistence in the developer forums. Losing an instance due to a failed or incorrect reboot is catastrophe for most computing environments. Obviously making gratuitous use of S3 for backups and write-ahead logging is key to any successful EC2 enterprise deployment. Is there ever going to be a point where an image can be considered persistent? PS: Postgresql seems to win hands down over MySQL in this respect; WAL is trivial to implement with Postgresql).
2) Dynamic IP is a show stopper for any mission critical application. When will static IPs be available?
3) No custom kernel is a show stopper. For any large scale clustering operation (openMosix, MPI, PVM etc) a custom kernel is a requirement. Your kernel image doesn't support any form of clustering, so Amazon's ability to sell to the scientific or educational community for clustering and/or supercomputing applications is severely limited in that regard.
4) Does EC2 guarantee any sort of throughput for intra-EC2 and S3, as well as EC2 instance-to-instance communications? Does subnet allocation and/or the current DHCP lease schema ever put an instance at a remote geographic location where round trip latency between instances would be reduced to WAN speeds instead of LAN connected speeds?
5) Load balancing? As the current IP address allocation schema many times issues instances on disparate broadcast domains, any layer-2 load balancing system breaks. Most commercial load balancing solutions are done at layer-2, which must be hosted on the same broadcast domain as the instances; the same goes for most open source load balancing packages.
6) Any 802.1q support planned for VLAN creation?
1) There are many references to "lost images" and non-persistence in the developer forums. Losing an instance due to a failed or incorrect reboot is catastrophe for most computing environments. Obviously making gratuitous use of S3 for backups and write-ahead logging is key to any successful EC2 enterprise deployment. Is there ever going to be a point where an image can be considered persistent? PS: Postgresql seems to win hands down over MySQL in this respect; WAL is trivial to implement with Postgresql).
2) Dynamic IP is a show stopper for any mission critical application. When will static IPs be available?
3) No custom kernel is a show stopper. For any large scale clustering operation (openMosix, MPI, PVM etc) a custom kernel is a requirement. Your kernel image doesn't support any form of clustering, so Amazon's ability to sell to the scientific or educational community for clustering and/or supercomputing applications is severely limited in that regard.
4) Does EC2 guarantee any sort of throughput for intra-EC2 and S3, as well as EC2 instance-to-instance communications? Does subnet allocation and/or the current DHCP lease schema ever put an instance at a remote geographic location where round trip latency between instances would be reduced to WAN speeds instead of LAN connected speeds?
5) Load balancing? As the current IP address allocation schema many times issues instances on disparate broadcast domains, any layer-2 load balancing system breaks. Most commercial load balancing solutions are done at layer-2, which must be hosted on the same broadcast domain as the instances; the same goes for most open source load balancing packages.
6) Any 802.1q support planned for VLAN creation?
Footcow said,
Wrote on January 7, 2007 @ 9:29 pm
It is time to realize that Postresql is more robust and more effective than Mysql !
It is like comparing a sports car out of toy and a true car!
CodeWord: Apokalyptik said,
Wrote on January 8, 2007 @ 3:23 pm
[…] I have written mostly about MySQL here in the past. The reason for this is simple: MySQL is what I know. I have always been a die hard “everything in its place and a place for everything” fanatic. I’ll bash Microsoft with the best of them, but I still recognize their place in the market. And now it’s time for me to examine the idea of PostgreSQL And this blog entry about amazon web services is the reason. I don’t claim to exactly agree with everything said here… as a matter of fact I tend to disagree with a lot of it… but I saw “PS: Postgresql seems to win hands down over MySQL in this respect; WAL is trivial to implement with Postgresql)” and thought to myself: “hmm, whats that?” I found the answer in the PostgreSQL documentation on Write Ahead Logging (WAL) and it all made sense! The specific end goal here is Continuous Archiving and Point-In-Time Recovery (PITR). This plus the S3 Infinidisk certainly do make for an interesting concept. One that I am eager to try out! I imagine that the community version of infinidisk would suffice here since we’re not depending on random access here… that ought to make for some chewy goodness! […]
purewebdev said,
Wrote on January 8, 2007 @ 5:02 pm
Personally I’m a big fan of MySQL, and it might also be because I haven’t needed any of the advanced features that other relational databases like DB2 or Oracle provide. Not being a dba myself, basic inserting, editing, and deleting of records is about all I need, but I need it done well. In this respect, mysql has performed very well for me, and I can’t complain. MySQL is growing and more features are being added, whatever it doesn’t do I can always do in php, so I don’t think I’ll be looking at Postgresql anytime soon.