1

  1. Followed the instruction of http://docs.vespa.ai/documentation/vespa-quick-start.html and issued yql-like curl (curl -s http://localhost:8080/search/?yql=select%20%2A%20from%20sources%20%2A%3B), got the error msg as follows, "message": "Could not instantiate query from YQL", could anyone point out if I missed anything to start any service?

  2. I want to store all the documents in physical memory for fast query, is there any configuration for me to achieve that? btw, is the doc compressed by default? (Also, I'd like to avoid disk io when feeding documents)

  3. Appreciate if anyone could share some internal architecture design doc for content/search node, thanks.

//// 1 works by comment #1.

4

3 回答 3

2

我会让其他人回应第 2 点和第 3 点，但我对第 1 点的猜测是您错过了 yql 查询中的“where”子句，因此失败。

于 2018-03-28T08:51:28.257 回答
2

3）我们没有更新的设计文档，很抱歉。如果您提出更具体的问题，我可以为您提供答案或找到一些东西（但我想您已经在https://github.com/vespa-engine/vespa/issues/5434中做到了）。

于 2018-04-04T13:53:39.243 回答
1

在 2) 上，您可以将字段定义为属性，并拥有一个仅引用属性字段的自定义文档摘要。见http://docs.vespa.ai/documentation/attributes.html http://docs.vespa.ai/documentation/document-summaries.html &

于 2018-03-28T10:24:31.447 回答