This document provides instructions for compiling and testing the WDT (Facebook's open-source data transfer library). It summarizes the steps to install prerequisites like Cmake and OpenSSL. It then describes compiling WDT from source, including issues encountered with specific library versions. The document tests WDT's transfer speed compared to SCP by sending a 5GB directory from one Ubuntu system to another. It notes WDT requires specifying a start port unlike SCP.
This document provides instructions for compiling and testing the WDT (Facebook's open-source data transfer library). It summarizes the steps to install prerequisites like Cmake and OpenSSL. It then describes compiling WDT from source, including issues encountered with specific library versions. The document tests WDT's transfer speed compared to SCP by sending a 5GB directory from one Ubuntu system to another. It notes WDT requires specifying a start port unlike SCP.
The document compares the performance of MySQL with the InnoDB and TokuDB storage engines on a system with high-performance SSD storage. It finds that InnoDB has better performance than TokuDB for most workloads tested, especially when thread counts increase. However, TokuDB provides much better data compression, nearly halving the storage size compared to InnoDB. For systems with very fast storage, InnoDB's compression is not as beneficial, but its performance is still superior in most tests.
The document discusses how NVMFS Benchmark replaces traditional double-write operations with single atomic writes to improve MySQL performance on flash storage. It provides details on installing and configuring NVMFS and MariaDB for atomic writes. Benchmark results show NVMFS delivers 20% higher TPS and 25% higher QPS compared to Ext4. The document questions a SanDisk benchmark that showed Ext4 outperforming NVMFS and seeks more details on the SanDisk test.
The document summarizes new features in MySQL 5.7, including improvements to InnoDB performance for read-only and read-write workloads, faster connection handling, bulk data load improvements, statement timeouts, multiple user level locks, and other features to improve scalability, concurrency, and performance.
This document provides 10 tips for optimizing MySQL database performance at the operating system level. The tips include using SSDs instead of HDDs for faster I/O, allocating large amounts of memory, avoiding swap space, keeping the MySQL version up to date, using file systems without barriers, configuring RAID cards for write-back caching, and leveraging huge pages. Overall, the tips aim to improve I/O speeds and memory usage to enhance MySQL query processing performance.
MySQL async message subscription platformLouis liu
?
The document discusses Roma, an asynchronous MySQL message system that subscribes to messages from the Canal system. Roma stores these messages in its own storage (MetaQ) since Canal cannot reliably store messages. Clients can then subscribe to the messages from Roma. Roma acts as an intermediary between Canal and clients to ensure messages are reliably stored.
The document discusses performance testing of HBase using YCSB (Yahoo! Cloud Serving Benchmark). It details steps taken to build and configure YCSB to work with HBase on a CDH cluster, load 50 million records into an HBase table, and run YCSB workload tests in modes A-K with threading from 1 to 128. Graphs show results for throughput, latency for each workload and threading configuration. The HBase environment consisted of 3 region servers and 10 data nodes on CentOS 6.5 servers with a total of 990GB memory and 24 CPU cores across servers.
YCSB is a tool for evaluating NoSQL databases. It provides various ways to simulate different types of workloads and loads on NoSQL databases. It can comprehensively test databases before they go live. The document discusses using YCSB to load data into MongoDB and run different workloads, such as read-heavy, write-heavy, and read-modify-write, and analyzes the results using metrics like average latency, throughput, and error rates.
MySQL Fabric is a clustering solution for MySQL introduced by Oracle after acquiring MySQL. It can provide high availability (HA) capabilities for MySQL through clustering and can also provide data sharding functionality. The HA mode relies on MySQL 5.6's GTID replication capability, while data sharding can be implemented using techniques like range-based or hash-based sharding. Fabric demonstrates Oracle's commitment to advancing MySQL towards mature enterprise-grade applications.
The document summarizes new features in MySQL 5.5 and 5.6. Some key points:
- MySQL 5.5 improved InnoDB performance, added new monitoring tools, and supported features like multi-buffer pools.
- MySQL 5.6 focused on improvements to replication like GTIDs for easier management, multi-threaded slaves for performance, and crash-safe replication.
- Other new features included online DDL support and transportable InnoDB tables to move data between servers.
This document discusses log-structured file systems and their relevance to solid state drives (SSDs). It explains that the core assumption of log-structured file systems, that writes are expensive and reads are cheap, is true for the basic components of SSDs, NAND flash memory. Writes to flash must be done in large blocks while reads can occur in smaller granularities. The document also describes garbage collection in SSDs, how data becomes invalid as it is rewritten, and how the garbage collection process works to free up flash blocks.
Yihaodian is a Chinese online grocery business founded in 2008. It operates a B2C e-commerce website and "virtual stores" in public areas that allow customers to scan codes and purchase groceries online. In 2011, Walmart first invested in Yihaodian and gained a 51% controlling stake by 2012. The document discusses Yihaodian's IT architecture, including its use of cloud platforms, Hadoop clusters, Oracle databases, real-time computing systems, MySQL and MongoDB databases distributed across multiple servers and data centers.
4. Jk.cn
route add -host 10.128.7.121 gw 10.128.6.84
route add -host 10.128.7.122 gw 10.128.6.84
在 test server 上将线上(online server)上的 mysql client IP 请求全部路由到 assistant server 由 assistant sever 跟 这些 client 机器完
成握手,从而完成整个过程。Test server 本身并不负责这些连接的建立。