PhonePe Case Study - Detailed

Customer Case Study
How PhonePe/Walmart Scaled

By 2000% and Saved $5 Million
With Acceldata
Case Study: PhonePe
Problem: Results (>) Large HBase OLTP

Scaling and performance issues on • Scaled data infrastructure by 2000% – 70- Cluster
open-source OLTP and OLAP. 1500 nodes – to support massive business
expansion (>) Hadoop, HBase,
Solution: • Delivered 99.97% availability across its Big
Hive, Spark,
Acceldata Pulse helped PhonePe monitor Data infrastructure Ranger, Kafka
HBase, Spark, and Kafka to distinguish
between infrastructure issues and seasonal
• Saved $5m+ in annual software licensing
(>) 400M cash
costs
and campaign-based anomalies. Transactions/
• Improved engineering productivity by
month
minimizing daily emergencies
(>) Open-source,
HDP binaries
acceldata.io 2
PhonePe is a Walmart subsidiary that provides more than 350 million consumers across India
with the ability to send and receive money, make payments at more than ten million physical
and online retail stores, use ATMs and invest in mutual funds and other securities.
PhonePe’s Challenge:
PhonePe uses a variety of open source data technologies, such as adopting the service, all while adding Hive LLAP, Spark 3.x and Druid
Apache Hbase, HDFS, Kafka, Spark, and Spark Streaming, to run their to the platform, technologies that were needed to support new
high-volume, real-time payments and cash transfer platform. With products and business requirements.
hundreds of millions of customers and millions of merchants on
the system, PhonePe’s Data Warehouse cluster must be highly Even in the early stages of this infrastructure expansion, the
performant, reliable and transparent, which includes the ability to technology team experienced tremendous pressure on system
accurately report on system and business performance to internal performance and reliability. Key engineers spent the majority of their
and external stakeholders 24/7. time firefighting problems and searching for causes behind data
application issues and infrastructure failure instead of focusing on
Scaling to Meet The Needs of Growing increasing scale and new capabilities as required by the business.
Data Infrastructure
PhonePe’s Chief Reliability Officer, Burzin Engineer, quickly realized
As PhonePe’s business grew explosively in 2018-19, the company that his team needed tools to improve visibility into every aspect of
embarked on a massive data infrastructure expansion in terms of the company’s data operations. WIthout more advanced tools that
both scale and new technologies. The company needed to increase matched the sophistication of his core open source technologies,
the size of its Hadoop infrastructure to support tens of millions of PhonePe’s critical data initiative would fail, jeopardizing the
new consumers and millions of new merchants who were rapidly company’s growth prospects and business success.
acceldata.io 3
The Acceldata Solution:
After gaining an understanding of PhonePe’s objectives and challenges
with Burzin Engineer and the PhonePe team, Acceldata demonstrated
how its Pulse data observability tool could provide real-time monitoring
of Hbase, Hive, and Spark data pipelines.
Acceldata Began Delivering Value in 24 Hours
Related image, icon motif or Acceldata

The PhonePe team implemented Pulse in less than a day and immediately
Diagram could go in this space.
began to identify problems with HBase region servers and tables that “Acceldata supports our hyper-growth and
were under pressure. Pulse helped PhonePe distinguish between HBase helps us manage one of the world’s largest
cluster issues caused by hardware or poorly designed tables and instant payment systems. PhonePe’s
anomalies resulting from seasonal and campaign-related surges. biggest-ever data infrastructure initiative
would never have been possible without
PhonePe had previously tried to use open-source and other commercially
Acceldata.”
available tools, like HBase Console and Ambari in addition to building
single metric grafana dashboards, for root cause analysis but found Burzin Engineer
that they were insufficient. HBase Console, for example, only provided Founder & Chief Reliability Engineer
aggregated information and required significant time and analysis from
highly experienced data engineers before it delivered useful intelligence.
In contrast, Pulse directs users to the problem’s root cause quickly and
clearly through automated alerts and easy-to-read dashboards. In
many cases, Pulse even recommends fixes to solve the problem.
acceldata.io 4
Results: Multi-Dimensional Data
In the first 18 months of using Acceldata Pulse, PhonePe has Observability
been able to realize these, among other, benefits:
(>) Scale data infrastructure rapidly from 70 to more than 1500 Enterprises are overwhelmed with the challenges
Hadoop nodes; more than 2000% growth of observing, operating, and optimizing large-scale
data systems.
(>) Deliver 99.97% availability across its Hadoop infrastructure
Multi-dimensional data observability can simplify
(>) Eliminate day-to-day engineering involvement and firefighting modern data pipelines by monitoring and correlating
on outages and performance degradation issues data workload events across application, data, and
infrastructure layers to resolve issues that break
(>) Support multi-cluster data and workload management with production analytics and AI workloads.
uniform configurations
The right data observability tools can significantly
(>) Upgrade systems and migrate to new applications and nodes improve enterprise data system performance, cost,
with no performance degradation and agility.
(>) Reduce data warehouse costs by 65%, while eliminating the

need for expensive commercial data warehousing licenses Try Acceldata - The Only Multi-Dimensional
Data Observability Cloud
acceldata.io 5

PhonePe Case Study - Detailed

Uploaded by

Copyright:

Available Formats

PhonePe Case Study - Detailed

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PhonePe Case Study - Detailed

Uploaded by

Copyright:

Available Formats

Customer Case Study

How PhonePe/Walmart Scaled

Problem: Results (>) Large HBase OLTP

Acceldata Began Delivering Value in 24 Hours

Related image, icon motif or Acceldata

(>) Reduce data warehouse costs by 65%, while eliminating the

You might also like