TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Show HN: TiloDB – serverless entity resolution technology

19 点作者 Major_Grooves将近 4 年前

4 条评论

nathanappere将近 4 年前
Interesting use case.<p>Some quick feedback on the homepage: - &quot;predictable&quot;: serverless is not exactly &quot;predictable&quot;. On the other end, having a set number of instances is :) - &quot;low cost&quot;: this assumes specific patterns of usage under a threshold where it&#x27;s profitable to use serverless. That&#x27;s entirely &quot;use-case dependant&quot; &amp; a bit misleading. - &quot;constant speed&quot;: so you&#x27;re saying that the worst case scenario is bounded to 150ms, no matter the data set?<p>Pointing these out because it makes it read as &quot;yet another DB vendor with some unrealistic claims&quot; :)
评论 #27515922 未加载
Major_Grooves将近 4 年前
We were the tech team at a European consumer credit bureau. We could not find any technology that could handle our scale of data with the required speed, efficiency and cost. So we spent the last three years developing a new entity resolution technology that we call TiloDB.<p>We think TiloDB has lots of potential outside this credit bureau, so hope to release the technology as open source software in the near future.<p>We would welcome your comments, thoughts or potential use cases. In the article you will find a description of the technical challenge we faced and how we solved it. There is also an interactive demo so you can play with the entity resolution yourself, and see the livestream of data submitted by other people.
yamalight将近 4 年前
That looks pretty neat, although not all things are exactly clear.<p>1. You claim that existing graph databases were not fast enough - do you have any benchmark data that compares them with your solution on given dataset?<p>2. From the description - it seems like you are focusing purely on Person type of data - is that correct? Or is that just the first use case &#x2F; demo?<p>3. Do you support more advanced query langs, e.g. SPARQL?<p>edit: formatting
评论 #27516406 未加载
ewe将近 4 年前
Neat! I can see few use cases, where it would make sense in our case. Is it publicly available?
评论 #27517085 未加载
评论 #27517072 未加载