[
https://issues.apache.org/jira/browse/HDFS-7285?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14302591#comment-14302591
]
Zhe Zhang commented on HDFS-7285:
---------------------------------
bq. I think it's incorrect. For example, we have a file, and it's length is
128M. If we use 6+3 schema, and ec stripe cell size is 64K, then we need
(128*1024K)/(6*64K) = 342 block groups.
Aah I see where the confusion came from. Sorry that the design doc didn't
explain clearly the different parameters. When the client writes to a striped
file, the following 3 events happen:
# Once the client accumulates 6*64KB data, it does _not_ flush the data to the
DNs. The client buffers the data and starts buffering the next 6*64KB stripe.
# Once the client accumulates {{1024 / 64 = 16}} stripes -- that is 1MB for
each DN -- it flushes out the data to DNs.
# Once the data flushed to each DN reaches 128MB -- that is {{128MB * 6 =
768MB}} data overall -- it allocates a *new block group* from NN.
Section 2.1 of the QFS [paper |
http://www.vldb.org/pvldb/vol6/p1092-ovsiannikov.pdf] has a pretty detailed
explanation too.
> Erasure Coding Support inside HDFS
> ----------------------------------
>
> Key: HDFS-7285
> URL: https://issues.apache.org/jira/browse/HDFS-7285
> Project: Hadoop HDFS
> Issue Type: New Feature
> Reporter: Weihua Jiang
> Assignee: Zhe Zhang
> Attachments: ECAnalyzer.py, ECParser.py,
> HDFSErasureCodingDesign-20141028.pdf, HDFSErasureCodingDesign-20141217.pdf,
> fsimage-analysis-20150105.pdf
>
>
> Erasure Coding (EC) can greatly reduce the storage overhead without sacrifice
> of data reliability, comparing to the existing HDFS 3-replica approach. For
> example, if we use a 10+4 Reed Solomon coding, we can allow loss of 4 blocks,
> with storage overhead only being 40%. This makes EC a quite attractive
> alternative for big data storage, particularly for cold data.
> Facebook had a related open source project called HDFS-RAID. It used to be
> one of the contribute packages in HDFS but had been removed since Hadoop 2.0
> for maintain reason. The drawbacks are: 1) it is on top of HDFS and depends
> on MapReduce to do encoding and decoding tasks; 2) it can only be used for
> cold files that are intended not to be appended anymore; 3) the pure Java EC
> coding implementation is extremely slow in practical use. Due to these, it
> might not be a good idea to just bring HDFS-RAID back.
> We (Intel and Cloudera) are working on a design to build EC into HDFS that
> gets rid of any external dependencies, makes it self-contained and
> independently maintained. This design lays the EC feature on the storage type
> support and considers compatible with existing HDFS features like caching,
> snapshot, encryption, high availability and etc. This design will also
> support different EC coding schemes, implementations and policies for
> different deployment scenarios. By utilizing advanced libraries (e.g. Intel
> ISA-L library), an implementation can greatly improve the performance of EC
> encoding/decoding and makes the EC solution even more attractive. We will
> post the design document soon.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
