kafka across data centers

InfiniBand now plays a key role in high-speed network architecture technology. By aggregating computing nodes in a computer cluster to accelerate computing performance, InfiniBand has made great strides in the field of high-performance computing. Although the InfiniBand Network performs better in terms of high bandwidth, low latency, and overall cost efficiency, there is still a long way to go for the widely used InfiniBand Network. But as my colleague Mike Matchett mentioned in a recent report,

HVAC: technical principle of energy-saving air conditioners in data centers According to statistics, air conditioning consumes 20% of the total energy consumption in the communication room ~ 45%, and some may even exceed 60%. Therefore, to ensure the normal operation of communication equipment, energy-saving should first begin with the large energy-consuming user of air conditioners. Currently, most of the

MPO/MTP high-density optical fiber pre-connection system is mainly used in three fields: application of high-density environment in data centers, application of optical fiber to building, application of optical fiber to optical splitter, 40G, 100g sfp, connection applications of SFP + and other optical transceiver devices. This article focuses on why MPO/MTP high-density optical fiber pre-connection system

Cabling slot design in data centers In the data center, all devices are interconnected through optical fiber cables or cables. Therefore, when you enter the data center, you will see various cables connected to different devices. How to wiring so many cables is a learning. The quality of wiring not only affects the ap

Good grounding of metal pipes can reduce interference and improve the line fire protection level of the equipment room. At the same time, the metal cable trough is used as the routing material to make full use of the cable slot's good scalability and easily add the characteristics of cables. Generally, the cable slots are arranged around the devices. Currently, the parallel layout of cabinets and into racks is more and more common. Generally, a cable slot is arranged for each cabinet, or a cable

Green cloud computing has many benefits: lower costs, faster upgrades, and simpler O M. From the perspective of social interests, it reduces the environmental burden by reducing carbon emissions. "The energy use of data centers is an important consideration. Cloud solutions are more efficient than traditional systems ." Eric Woods of Pike Research said. This is because the carbon footprint of IT has recent

Cannot be overly obsessed with virtualized data centers At present, the industry's interest in data center and virtual environment automation is growing-automation frees administrators from daily hard work, IT professionals are allowed to focus on more strategic projects that can bring long-term practical benefits to the business. However, virtualization automat

modern cooling system HVAC systems are typically the second largest power consumption system in the data center, with data center Users Group Estimating 37% of the energy consumed in the datacenter. Air conditioning system can only achieve high efficiency when dealing with a small range, in the face of large-scale data center, most cooling system cooling task c

How to better monitor data centers Enterprises hope that before the problem arises, the intelligent analysis function can bring the enterprise's system monitoring to a new level. Now, IT is an important capability for the IT department to make proper planning of production capacity to maintain smooth operation. In addition to achieving IT and business goals, IT can help maintain the IT infrastructure of en

The reliability, delay and consistency of the distributed system are general problems, not limited to the database, and Cassandra provides a good solution to the problem.Cassandra claims to be able to achieve the efficient access to database access across data centers, and it is implemented in a way that gives users the tradeoff between latency, throughput, and consistency. Cassandra provides two levels of

Construction solution for local data centers with 150 nodes I. Local Data Center When I came into the company last year, there were only 20 people, and now it has grown to 100. At that time, the office was quite small and could accommodate up to 40 employees. By March, the company plans to move its office. In view of the office environment at that time, in fact,

to vultr.com SSD drive, this 230mb/s effect is actually very good, you have to know that this is the nature of the storage of VPS.Netherlands (Amsterdam ) Room: is still 1000M port, it is obvious that two NL node test a lot, Holland to Holland, 460M I/O for the general toss Emperor is OK, and this is E3 series of CPU, the largest 32G memory support die, Overall, reliable.Atlanta, USA:Chicago, USA:DALLAS, USA Room:France Room:German room:UK Room:LOS Angeles, USA:Miami, USA:New Jersey, USA:Seattl

messages. How to ensure the correct consumption of messages. These are the issues that need to be considered. First of all, this paper starts from the framework of Kafka, first understand the basic principle of the next Kafka, then through the KAKFA storage mechanism, replication principle, synchronization principle, reliability and durability assurance, and so on, the reliability is analyzed, finally thro

Original link: http://www.ibm.com/developerworks/cn/opensource/os-cn-spark-practice2/index.html?ca=drs-utm_source= Tuicool IntroductionIn many areas, such as the stock market trend analysis, meteorological data monitoring, website user behavior analysis, because of the rapid data generation, real-time, strong data, so it is difficult to unify the collection and s

Personal opinion: Big data we all know about Hadoop, but not all of it. How do we build a large database project. For offline processing, Hadoop is still more appropriate, but for real-time, relatively strong, the amount of data is large, we can use storm, then storm and what technology collocation, to be able to do a suitable project. We can refer to the following.You can read this article with the followi

Http://www.aboutyun.com/thread-6855-1-1.htmlPersonal opinion: Big data we all know about Hadoop, but not all of it. How do we build a large database project. For offline processing, Hadoop is still more appropriate, but for real-time, relatively strong, the amount of data is large, we can use storm, then storm and what technology collocation, to be able to do a suitable project. We can refer to the followin

First of all, this is my original article, but also refer to the network of the Great God's articles plus their own summary, welcome to the Great God pointed out the mistake! We make progress together. Where the 1.kafka data exchange is done. Kafka is designed to make every effort to complete data exchange in memory,

of various data senders in the log system and collects data, while Flume provides simple processing of data and writes to various data recipients (customizable) capabilities. typical architecture for flume:flume data source and output mode:Flume provides 2 modes from consol