Skip to main content

Hadoop world and SAP Shops coming closer

With some major announcements done recently on partnerships between SAP HANA and Hadoop distributions, there is now a formal channel for Hadoop to make inroads in SAP shops. The 'SAP shop' is an informal lingo term used for captive enterprise organizations which utilize SAP for ERP, SCM, CRM, HR and other enterprise needs. These announcements are probably one of the biggest enterprise push to Hadoop this year besides EMC’s Hawq and IBM’s PureData Systems blitz in the earlier quarters.

Symbiotic relationship

Within the startup and internet companies world, Hadoop has been used with in-memory technologies in a wide variety of integrated ways which include Redis and Memcached. Certain organizations have used variant architecture where HDFS file system has been used for data storage, Redis for data cache and  database like MongoDB for key-value store. However, Redis has still some way to go for enterprise adoption besides current constraints on scalability. Similarly, Memcached has had its own limitations like data loss culpability on node/process restart. As a symbiotic mutual exchange reward, it is probably this lacuna that SAP HANA aims to exploit within the ardent followers of Hadoop in internet companies.  Although not open source or free, SAP HANA has also alongside moved beyond license based usage to cloud based billing and probably can be a bit more appealing to the cost conscious startup world.

Enterprise Architecture decisions

Beyond the initial hype and confusion which concerned the replacement of RDBMS with Hadoop or replacement of Hadoop with SAP HANA, there is much larger sense of consensus and clarity within enterprise adopters on co-existence of RDBMS, NoSQL DB, Hadoop and SAP HANA. It is now largely accepted that pre-defined schema data fits better in DB which may be relational, NoSQL or in-memory. On the other hand, if the data is largely unstructured and it is better left to programmer or analyst to explore the data, Hadoop may be a better choice. Similarly, data in PetaByte or ExaByte range makes a natural fitment to Hadoop while data less than 5TB would better fit in a DB.  One would need to keep in mind that the 5TB range includes metadata, temporary table and buffer size.

Price point for lower data volumes makes an appealing cost case for Hadoop while in the higher volume range, you may have to take a prudent call on either of the technologies. Skills is not as much of an issue any longer since most organizations prefer to train their resources and with a period of time, there will be a decent number of folks on the winning products among these technologies. Similarly, within the enterprise architecture, if the organization is using multiple vendor systems for SCM, ERP, payroll, campaign management etc, there is more than a likely chance, the organization would like to consolidate its data in a central warehouse or a Hadoop driven repository.

SAP offering

SAP Real-Time DataPlatform now combines Hadoop with following SAP technologies like:
-          SAP HANA – in memory DB for faster access;
-          SAP Sybase IQ- for ‘federated’ queries by considering HDFS/Hive as external DB and joining results in Sybase;
-          SAP Data Services – for data interactions using HiveQL or Pig;
-          SAP Information Steward – for improving data quality;
-          SAP Sybase Event Stream Processor – real time analysis of events & data and pushing required data to HDFS;
-          SAP BusinessObjects, SAP BusinessObjects Explorer and SAP Crystal Reports – for reporting, visualization and building data models.

We will keep a watch on enterprise success stories but for now, the merging of both worlds seems to sound like a win-win at both ends.

Comments

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 …