Skip to main content

10 parameters for Big Data networks

Big Data and Hadoop clusters involve heavy volume of data and in many instances high velocity in bursty traffic patterns. With these clusters finding in-roads in enterprise data centers, the network designers have a few more requirements to take care. Listed below are 10 parameters to evaluate while designing a network for Big Data and Hadoop cluster.

10) Available and resilient

- Allows network designs with multiple redundant paths between the data nodes than having one or two points of failure.
- Supports upgrades without any disruption to the data nodes

9) Predictable

- Right sizing the network configuration (1GbE/10GbE/100GbE switch capacity) to achieve predictable latency in network
- real time latency may not be required for batch processing

8) Holistic network

- one network can support all workloads : Hadoop, NoSQL, Warehouse, ETL, Web
- support Hadoop and existing storage systems like DAS, SAN, or NAS

7) Multitenant

- be able to consolidate and centralize Big Data projects
- have capability to leverage the fabric across multiple use cases

6) Network partitioning

- support separate Big Data infrastructure from other IT resources on the network
- support privacy and regulatory norms

5) Scale Out

- provide seamless transition as projects increase in size and number
- accommodate new traffic patterns and larger, more complex workloads

4) Converged/ unified fabric network

-  target a flatter and converged network with Big Data as an additional configurable workload
- provide virtual chassis architecture with provision to logically manage access to multiple switches as a single device

3) Network intelligence

- carry any-to-any traffic flows of a Big Data as well as traditional cluster over an Ethernet connection
-  manage single network fabric irrespective of data requirements or storage design

2) Enough bandwidth for data node network

- provision data nodes with enough bandwidth for efficient job completion
- do cost/benefit trade-off on increasing data node uplinks

1) Support bursty traffic

- support loading files into HDFS which triggers replication of data blocks or writing mapper output files and lead to higher network use in a short period of time causing bursts of traffic in the network.
- provide optimal buffering in network devices to absorb bursts


Popular posts from this blog

Low latency SQL querying on HBase

HBase has emerged as one of the most popular NoSQL database offering distributed, versioned, non-relational tables hosted on commodity hardware. However, with a large set of users coming from a relational SQL world, it made sense to bring the SQL back in this NoSQL. With Apache Phoenix, database professionals get a convenient way to query HBase through SQL in a fast and efficient manner. Continuing our discussion with James Taylor, the founder of Apache Phoenix, we focus on the functional aspects of Phoenix in this second part of interaction.
Although Apache Phoenix started off with distinct low latency advantage, have the other options like Hive/Impala (integrated with HBase) caught up in terms of performance?
No, these other tools such as Hive and Impala have not invested in improving performance against HBase data, so if anything, Phoenix's advantage has only gotten bigger as our performance improves.  See this link for comparison of Apache Phoenix with Apache Hive and Cloudera Im…

Data deduplication tactics with HDFS and MapReduce

As the amount of data continues to grow exponentially, there has been increased focus on stored data reduction methods. Data compression, single instance store and data deduplication are among the common techniques employed for stored data reduction.
Deduplication often refers to elimination of redundant subfiles (also known as chunks, blocks, or extents). Unlike compression, data is not changed and eliminates storage capacity for identical data. Data deduplication offers significant advantage in terms of reduction in storage, network bandwidth and promises increased scalability.
From a simplistic use case perspective, we can see application in removing duplicates in Call Detail Record (CDR) for a Telecom carrier. Similarly, we may apply the technique to optimize on network traffic carrying the same data packets.
Some of the common methods for data deduplication in storage architecture include hashing, binary comparison and delta differencing. In this post, we focus on how MapReduce and…

Pricing models for Hadoop products

A look at the various pricing models adopted by the vendors in the Hadoop ecosystem. While the pricing models are evolving in this rapid and dynamic market, listed below are some of the major variations utilized by companies in the sphere.
1) Per Node:Among the most common model, the node based pricing mechanism utilizes customized rules for determining pricing per node. This may be as straight forward as pricing per name node and data node or could have complex variants of pricing based on number of core processors utilized by the nodes in the cluster or per user license in case of applications.
2) Per TB:The data based pricing mechanism charges customer for license cost per TB of data. This model usually accounts non replicated data for computation of cost.
3) Subscription Support cost only:In this model, the vendor prefers to give away software for free but charges the customer for subscription support on a specified number of nodes. The support timings and level of support further …