简体   繁体   English

Apache Spark和Flink的能耗如何测量

[英]How to measure energy consumption of Apache Spark and Flink

How to measure energy consumption of Apache Spark and Flink are there any tool or technology to measure the energy consumption of Spark and Flink Apache Spark和Flink的能耗如何测量 有没有什么工具或者技术可以测量Spark和Flink的能耗

Welcome to stack overflow !欢迎来到堆栈溢出!
Apache spark and Flink are frameworks that are used for distributed computing and stream processing respectively.There is nothing like frameworks consuming computational power. Apache spark 和 Flink 分别是用于分布式计算和 stream 处理的框架。没有什么比框架更耗算力的了。

If I get your intention correct, you were asking about computational power consumption of applications which were developed using above frameworks.如果我理解你的意图是正确的,那么你是在询问使用上述框架开发的应用程序的计算功耗。

So, answering that, generally such applications are deployed in any any on the servers / cloud providers.因此,回答这个问题,通常此类应用程序部署在任何服务器/云提供商上。 These cloud providers give in built metrics to track the CPU / memory consumption.这些云提供商提供内置指标来跟踪 CPU / memory 消耗。

If you are running these applications on you local, you can use tools like jprofiler , jconsole , visualVM .如果您在本地运行这些应用程序,则可以使用 jprofiler 、 jconsolejprofilervisualVM These tools gives you visual representation of CPU consumption, memory consumption etc.这些工具为您提供 CPU 消耗、memory 消耗等的可视化表示。

PS:附言:
A software framework is an abstraction in which software providing generic functionality can be selectively changed by additional user-written code, thus providing application-specific software.软件框架是一种抽象,其中提供通用功能的软件可以通过附加的用户编写的代码进行选择性更改,从而提供特定于应用程序的软件。

Distributed computing,stream processing or say any computation can be implemented with out the above frameworks.分布式计算,stream 处理或者说任何计算都可以在没有上述框架的情况下实现。 But it will be chaos.但这将是混乱的。 Framework makes a dev's life easy.框架使开发人员的生活变得轻松。 There are trade offs coming along with ease due to over abstraction, that is a different discussion as a whole.由于过度抽象,容易权衡取舍,这是一个不同的整体讨论。

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM