You probably shouldn’t use DynamoDB

Edit: This article is firmly out of date now. I have not gone back to try dynamoDB but I have heard they have changed how throughput is calculated and mitigated this issue. So take what is below with a grain of salt! Avid readers of the Ravelin syslog will remember a story from last year about our use of DynamoDB. It outlined a few gotchas that we experienced and gave advice on different uses case

[soy-curd]

あなたはおそらくDynamoDBを使うべきではない

[avalon1982]

“Alternatives:... I will mention another managed solution that is far better than DynamoDB. Google Bigtable.”

[braitom]

DynamoDBの問題点について。この人の場合はGoogleのBigtableでは同じような問題は起きなかったとのこと。

[akulog]

気持ちはよくわかる

[yass]

" So if you request a write capacity of 1000 and you have 200 nodes what this really means is that each node only has a write capacity of 5. That is pitifully small and will cause throughput exceptions. "

[daikikohara]

こーゆーの書きたくなっちゃうくらい辛いのは分かる...