Mapreduce 1算法的缺點

Question

我想知道Map Reduce 1算法在MR2上的缺點。

Answer 1

Here are two exciting and significant additions to the Hadoop framework:  

• HDFS Federation: provides a name service that is both scalable and reliable. 

• YARN: Yet Another Resource Negotiator,it divides the two major functions  of the JobTracker(resource management and life cycle management) into separate   components.    

Hadoop 1.x的一個關鍵問題是提供了高度可用的名稱節點。 HDFS聯盟不僅提供HA名稱服務，還允許分配工作負載，因爲名稱節點現在可以水平擴展。

YARN提供了在Hadoop集羣中協商和執行作業的職責的邏輯分離。YARN的最終結果是一個新的，更通用的資源管理框架，它不僅僅適用於Map Reduce作業。

Here are some of the articles 
http://blog.cloudera.com/blog/2012/02/mapreduce-2-0-in-hadoop-0-23/ 
http://hortonworks.com/blog/introducing-apache-hadoop-yarn/ 

Hadoop 1.x is all about Map -reduce means you can run only map reduce but 
YARN is more general than MR and it should be possible to run other computing models like BSP besides MR. Prior to YARN, it required a separate cluster for MR, BSP and others. Now they they can coexist in a single cluster, which leads to higher usage of the cluster. Here are some of the applications ported to YARN. 

In the current system, JobTracker views the cluster as composed of nodes (managed by individual TaskTrackers) with distinct map slots and reduce slots, which are not fungible. Utilization issues occur because maps slots might be ‘full’ while reduce slots are empty (and vice-versa). Fixing this was necessary to ensure the entire system could be used to its maximum capacity for high utilization.. 

Also, it makes it possible to run different versions of Hadoop in the same cluster which is not possible with legacy MR, which makes is easy from a maintenance point. 

Answer 2

我覺得喜歡的是MR1的主要困難是

很難做到這一點需要全球性的，共享的狀態算法。