DB-EnginesInfluxDB: Focus on building software with an easy-to-use serverless, scalable time series platformEnglish
Informationen zu relationalen und NoSQL DatenbankmanagementsystemenEin Service von solid IT

Blog > Post

Evaluating Hbase and Aerospike Solutions for Real-time System of Record

von  Srini Srinivasan, Aerospike (sponsor) , 17. März 2020
Tags: Aerospike, NoSQL

This is a sponsored post.

Legacy platforms often cannot meet real-time demands needed for innovation: slower response times cost the business and its customers which leads to bad user experiences. Plus, about a third of companies have failed on big data initiatives because they were too complex, required a large budget, demanded more skills and were not agile enough.

More businesses today – no matter the industry or size – are demanding real-time data use in order to meet the innovation and growth demands necessary for survival in today's competitive marketplace.

The problem, however, is that many are faced with legacy platforms that can't keep pace. Data can't be leveraged quickly and slower response times cost the business and its customers in terms of slower response times that lead to a bad user experience.

Organizations face other headaches. For example, about a third of companies have failed on big data initiatives because they were too complex, required a large budget, demanded more skills and were not agile enough. On top of this, we know that data governance and security and global privacy regulations are significant hurdles to clear, as well.

Organizations must address their data collection issues immediately

Companies are collecting more data that is critical for:

  • Customer insights
  • Business insights
  • Strategic initiatives 
  • Faster and more accurate decisions for related products.

Thus, having reliable, convenient and fast data collection is paramount.

For example, in retail, real-time data is necessary for attracting and retaining customers who want personalization with every interaction, whether it's offering them a discount based on their loyalty or recommending a product based on past purchases. Even manufacturing will find real-time data critical in order to predict machine failure and potential plant shutdowns before they happen. Financial institutions will need to process data in real-time to better understand the risk profiles of their positions or determine fraud scores for transactions as they are in flight.

Start with your SLAs in mind

When making decisions about real-time data, it makes more sense to think about your SLAs – what do you really need? Then, you can consider the technology necessary to support it.

Once this is done, then it's necessary to look at how you're currently handling real-time data. Many organizations find they've outgrown systems such as HBase. 

While using HBase to handle real-time workloads, organizations face issues such as:

  • Server sprawl
  • Difficult-to-maintain clusters
  • Unpredictability and difficulty in configuration and management

Where the Aerospike NoSQL data platform can help

Aerospike is the only NoSQL data platform that can provide both scalability and speed because of its unique architecture. For example, HBase is primarily designed for low latency random access to billions of records and for horizontal scale, but its performance lags its peers in the NoSQL space. 

Consider some of the advantages of Aerospike replacing HBase in enterprise environments:

  • High performance (low latency) at scale without any compromises. Our multithreaded parallel processing at the CPU – together with our patented Hybrid Memory Architecture designed for flash storage devices – uniquely delivers predictable high performance at scale. We can ensure linear scale up using our Smart Client architecture that ensures concurrent access to multiple nodes in a cluster for achieving the lowest possible latency. For distributed environments, our Cross Datacenter Replication (XDR) feature allows multiple geographically dispersed data centers to stay in sync through high performance replication. HBase can't match the same performance at scale because  its architecture is not tuned to this kind of environment. Because HBase runs on HDFS, its Write-Ahead Logging (required for strong consistency) and use of multiple layers of storage makes the latency much worse than what is required by modern real-time applications.
  • Easy to configure, scale and manage. Aerospike is easy to install and uses the native storage system directly. Also, from the management perspective, most functions are automated. We support self-forming and self-healing clusters that are both rack-aware and data center-aware. There is no downtime required to a+dd or remove nodes from a cluster. We automatically redistribute partitions of data to ensure new resources are efficiently used. The rack awareness also means that data is correctly separated to allow high availability of the data even in the event of an entire data center going down, while still maintaining strong consistency. HBase has a significant complexity both during the configuration and the operation. In order to configure and manage HBase, users must understand and manage three different server types (Master Server, Region Servers, Zookeeper).
  • Fully supported Enterprise Edition NoSQL database. Our Enterprise Edition provides access to tested and certified builds, hot patches and 24 /7 Enterprise support. We are available on premises and in various cloud environments. With HBase, customers have to rely on distributors or open source. Technically, there is no Enterprise Edition for HBase: the Master is Open Source. 
  • Best HW utilization and operational simplicity. Generally, our users routinely save five times or more in TCO when compared with alternate NoSQL solutions. It's well known in the industry that HBase is expensive to implement and maintain for real-time workloads, in part because it requires lots of memory and high staffing, license and maintenance fees.

Companies such as Wayfair, and InMobi have moved from HBase to Aerospike. As more companies seek strong consistency, high throughput, low latency and a system that offers them the ability to be innovative and use their data in real time, we believe Aerospike can provide the kind of predictable performance and operational simplicity they need.

To learn more, watch our webinar:

Aerospike Webinar


About the Author:
Srini V Srinivasan

Srini V Srinivasan is a Founder and Chief Product Officer at Aerospike. When it comes to databases, Srini Srinivasan is one of the recognized pioneers of Silicon Valley. He has two decades of experience designing, developing and operating high scale infrastructures. He also has over a dozen patents in database, web, mobile and distributed systems technologies. Srini co-founded Aerospike to solve the scaling problems he experienced with Oracle databases while he was Senior Director of Engineering at Yahoo. Srini has a B.Tech, Computer Science from Indian Institute of Technology, Madras and both an M.S. and Ph.D. in Computer Science from the University of Wisconsin-Madison.

Teilen sie diese Seite mit ihrem Netzwerk

Featured Products

Datastax Astra logo

Bring all your data to Generative AI applications with vector search enabled by the most scalable
vector database available.
Try for Free

Neo4j logo

See for yourself how a graph database can make your life easier.
Use Neo4j online for free.

Milvus logo

Vector database designed for GenAI, fully equipped for enterprise implementation.
Try Managed Milvus for Free

Präsentieren Sie hier Ihr Produkt