Tag：Distributed
-
Time：2019-11-14
Author: aikesheng open source community background MYCAT has three fragmentation modes for data whose string type is fragmentation field: modulus hash, jumpstringhash and consistency hash Dble does not continue to use the consistency hash of MYCAT, except inheriting the modulus hash of MYCAT, but recommends the one with better performance and better balance.“jumpstringhash”Algorithm. introduce The […]
-
Time：2019-11-13
This article is from the public address TCTP, author TCTP. Original link: https://mp.weixin.qq.com/s/voer3wid1leor-o4sfptpa? Scene = 25 × wechat × redirect In 2018, our bank started an attempt based on ARM server platform. Currently, all cold standby data of tdsql are stored in CEPH storage cluster based on ARM server, which runs stably. In the context […]
-
Time：2019-11-13
On September 11, ant financial opened the elasticdl project at the 2019 Google Developer Conference Shanghai station, which is the industry’s first open-source system based on tensorflow to realize elastic deep learning. Open source address: elasticdl.org Kaiyuan China interviewed Wang Yi, project director of elasticdl, and gave a comprehensive introduction to the technical details of […]
-
Time：2019-11-12
On September 11, ant financial announced in Google Developer Day Shanghai 2019 that it had opened the distributed deep learning system elasticdl based on tensorflow 2.0 eager execution. As far as we know, elasticdl is the first deep learning system supporting elastic scheduling based on tensorflow. Wang Yi, the project leader, shared with us the […]
-
Time：2019-11-11
Dble Release Notes version 2.19.07.0 I. Introduction to dble project Dble is an enterprise level open-source distributed middleware, which is nicknamed “MYCAT plus” by people from the Jianghu. It is supported by the community with its simple, stable and continuous maintenance, good community environment and broad mass base; Dble official project: https://github.com/actiontech/dble If you are […]
-
Time：2019-11-11
Preface Table storagetablestore is a serverless NoSQL multi model database developed by Alibaba cloud for massive structured data storage. Tablestore has a variety of documents on Alibaba cloud’s official website, as well as many scenario and case articles, which are included in this collectionTable store authority guide. It is worth mentioning that the tablestore can […]
-
Time：2019-11-10
Nebula graph: an open source distributed graph database. As the only online graph database that can store trillions of nodes and edges with attributes, nebula graph can not only meet the requirements of millisecond level low latency query in high concurrency scenarios, but also achieve high service availability and ensure data security. The third issue […]
-
Time：2019-11-10
Recently, we expanded the tidb expression calculation framework and added the vectorization calculation interface. The initial performance test shows that most expression calculation performance can be greatly improved, and some of them can even be improved by 1-2 orders of magnitude. In order for all expressions to benefit, we need to implement vectorization for all […]
-
Time：2019-11-9
If I ask you how to deploy an etcd cluster on Google cloud or Alibaba cloud, you will automatically give the answer: of course, use etcd operator! In fact, almost overnight, kubernetes operator, a new thing, has become a de facto standard for developing and deploying distributed applications. Today, no matter etcd, tidb, redis, Kafka, […]
-
Time：2019-11-9
Abstract:Build an agile capability center and create the next generation of Digital Banking “operating system“! The little ant said: In recent years, more and more leading banks at home and abroad have devoted themselves to digital transformation. Digital transformation is the deep integration of technology and business model, and the final result is the change […]
-
Time：2019-11-8
At present, the Internet technology has already entered the stage of big data and artificial intelligence. Various computing models and computing engines emerge in endlessly. From MapReduce distributed computing 10 years ago, there will be a new computing engine change almost every three years, from Hadoop, the founder of the mountain, to storm, the later […]
-
Time：2019-11-7
Video link: https://developer.aliyun.com/live/1548? SPM = a2c6h. 12873581.0.0.71671566xoloy3z & group code = apachespark PPT information: https://www.slidestalk.com/alispark/sparkrelationalcache2019_ This sharing is mainly divided into the following four aspects: Project introduction technical analysis How to use performance analysis I. project introduction Project background Alibaba cloud EMR is an open-source big data solution. At present, many open-source components have been […]
