Altiscale, Inc., the leading provider of Hadoop-as-a-Service, today announced that Kerberos authentication is now available in the Altiscale Data Cloud. Altiscale is the first and only Hadoop-as-a-Service provider that will automatically include Kerberos authentication as part of its service offering.

Kerberos, originally developed at the Massachusetts Institute of Technology (MIT), is a user and network authentication protocol that leverages strong cryptography to help ensure that Hadoop users can access only the data that they are supposed to access, and not other, potentially sensitive data. Kerberos not only provides fortification against outside attackers, it also provides protection against malicious insider activity. While Kerberos is known to be very secure, it was not designed in the context of today’s highly dynamic, elastic cloud architectures. As a result, although Kerberos authentication is widely recognized as critical to Hadoop security, it has not been previously available in Hadoop-as-a-Service offerings. Altiscale’s automatically implemented Kerberos authentication will be unique in the marketplace.

“Altiscale is dedicated not only to providing high-performing Hadoop data environments, but also to providing technologies for data security and stewardship that satisfy the most demanding enterprise customers,” said David Chaiken, CTO of Altiscale. “Kerberos is a great addition to the full range of security technologies that are automatically provided in the Altiscale Data Cloud.”

“Kerberos authentication is foundational to a well-secured Hadoop environment,” said Joe Stein, Hadoop security expert and founder of the Big Data Open Source Security consultancy. “It is difficult to integrate Kerberos into cloud deployments of Hadoop, so it’s often not turned on, limiting the ability to create a true ‘data lake’ in the cloud.”

“To meet enterprise security requirements, Hadoop needs to be surrounded by a complete set of security technologies, policies, and procedures,” said Brian Schneirow, head of Security and Compliance for Altiscale. “For this reason, the experts at Altiscale have secured the Altiscale service with multiple, reinforcing layers of protection. In response to the evolving threat landscape and our understanding of new vulnerabilities, we are constantly strengthening our security program. Kerberos is one of many additions that Altiscale will release this year.”

Kerberos authentication in the Altiscale Data Cloud is now available for private beta and will be broadly available this spring.

In-person demonstrations will be available at the Strata + Hadoop World conference at the San Jose Convention Center from February 17 through February 20, 2015, Booth #929. Online demonstrations are available by request at https://www.altiscale.com/contact/

Altiscale announces Apache Spark on the Altiscale data cloud
Altiscale, Inc., the leading provider of Hadoop-as-a-Service, today announced that Apache Spark is now available on the Altiscale Data Cloud. Altiscale customers can now leverage Apache Spark on Apache Hadoop in order to achieve their critical analytical and business objectives. The addition of Apache Spark provides a broader array of analytical services for machine learning, stream processing, and data processing for large data sets.

“Altiscale is dedicated to helping customers quickly find value in the ever-increasing flood of data generated by the connected world,” said Raymie Stata, co-founder and CEO of Altiscale. “Apache Spark in the Altiscale Data Cloud ensures that customers can take advantage of the latest in-memory processing techniques as they are processing their data assets. The Altiscale Data Cloud is purpose-built to provide the fastest, most scalable Hadoop-as-a-Service and is the ideal place to run Spark.”

Apache Spark is an open source framework that is gaining adoption for its machine learning, interactive analytics, and streaming analytics capabilities for large datasets. Spark is appropriate for low-latency computations and iterative algorithms that employ its in-memory computing capabilities. At Altiscale, Spark is fully integrated into the larger Hadoop ecosystem, so customers benefit not only from Spark, but also from Hive, Pig, MapReduce, and even tools like R and H2O. All of these tools run side-by-side on the same Hadoop Data File System (HDFS) cluster, managed through YARN.

“There’s a common misunderstanding that you have to choose between Hadoop and Spark,” said David Chaiken, CTO, Altiscale. “Hadoop is a large ecosystem that includes storage, security, and multiple ways to process your data. Spark is a computing paradigm that fits into and runs best in the Hadoop ecosystem. At Altiscale, customers get the full benefit of leveraging Spark’s complementary strengths.”

Altiscale customers are already using both MapReduce and Apache Spark on the Altiscale Data Cloud. For example, one customer, who had already been using MapReduce to perform regular customer billing analysis upon tens of millions of customer records, had a need for fast, daily geographic analysis and reporting. Apache Spark was quickly added to their Altiscale Data Cloud subscription and the customer expects to expand its use.

Apache Spark runs reliably in the Altiscale Data Cloud, where all aspects of hardware, networking, security, software, tools, and operations are optimized for the processing and analysis of massive data sets. Learn more at https://www.altiscale.com/why-altiscale/

The Altiscale Data Cloud with Apache Spark is available now. Demonstrations of the Altiscale Data Cloud will be available in person at the Strata + Hadoop World conference at the San Jose Convention Center from February 17 through February 20, 2015, Booth #929. Web demonstrations are also available by request at https://www.altiscale.com/contact/