Big Lots and the path to optimal performance with OCI Observability & Management

January 2, 2024 | 6 minute read
Kathryn Jennings
Senior Principal Product Manager
This is a syndicated post, view the original post
Text Size 100%:

Guest Author – Satish Mukundachar, Director of Applications Development, Big Lots

Blog contributor - Vas Uppalapati - Master Principal Cloud Architect, Oracle

This blog focuses on Big Lots' adoption of Oracle Cloud Infrastructure (OCI). Mr. Satish Mukundachar, Director of Applications Development at Big Lots, has shared insights into their transition from dedicated infrastructure to OCI, highlighting key benefits they have achieved in the process.

 

Describe Big Lots’ business and your role at Big Lots.

Big Lots, Inc. is a prominent discount retailer with a widespread presence, operating over 1400 stores across the United States. The company is dedicated to its mission of enabling customers to "Live Big and Save Lots" by offering a diverse range of valuable items for their homes, including furniture, seasonal decorations, kitchenware, pet supplies, groceries, cleaning essentials, and more.

In my role as the Director of Applications Development, I oversee the management of two engineering teams. One of these teams is responsible for the design, development, and maintenance of www.biglots.com, the e-commerce platform that serves as Big Lots' online presence.

Our journey into the online realm began in 2016 when Big Lots had no e-commerce presence. We initiated the development of the Big Lots eCommerce website, which empowered customers to place orders for home delivery. Over time, we expanded our fulfillment options to include features such as Buy Online and Pickup in Store (BOPIS) and Same Day delivery, leveraging our vast network of 1400-plus stores.

Throughout this journey, we have consistently improved the customer experience by introducing personalization features and streamlining the order placement process, reducing friction for our online shoppers. Additionally, our website has become a valuable resource for customers researching significant purchases like furniture, providing customers with enhanced digital capabilities to facilitate their decision-making process.

What factors helped your decision to move to the Oracle Cloud?

From a technological perspective, our e-commerce website relies on Oracle's ATG Web Commerce framework. Initially, our e-commerce website was hosted on dedicated hardware, which posed certain limitations. Specifically, we encountered challenges in dynamically adjusting our resource allocation to match the fluctuating traffic patterns on the website.

In response to these limitations, we made the strategic decision to migrate to Oracle Cloud Infrastructure (OCI). This transition enabled us to leverage modern cloud capabilities, including vertical and horizontal scaling, advanced application monitoring, and enhanced security measures.

Additionally, the cost considerations played a significant role in motivating our move to the cloud. The migration to OCI yielded substantial cost savings for our organization, making it a highly beneficial choice.

How has Big Lots cloud adoption evolved since transitioning to the cloud?

During our cloud migration, we pursued a lift-and-shift approach, essentially replicating our existing infrastructure and mirroring what we had in our dedicated hardware setup.

Now, as we prepare for the upcoming holiday season, we are focused on enhancing our ability to scale both horizontally and vertically in response to our specific holiday demands which was a challenge with our previous dedicated hardware setup. We also intend to take advantage of the pay-as-you-go pricing model and make efforts to optimize resource utilization on the OCI Cloud.

Additionally, we have streamlined our Application Performance Monitoring (APM) to capitalize on cost-effective monitoring solutions and have reallocated resources to more strategic areas. With the assistance of the Oracle team, we've made necessary adjustments to ensure our systems are in a manageable and secure state as we approach the holiday season. Today our OCI solution includes Application Performance Monitoring, Stack Monitoring, and Logging Analytics.

How does Big Lots anticipate using Cloud capabilities as you grow and evolve your solution?

We aim to harness the full range of OCI DevOps capabilities and take advantage of the Synthetic monitoring tool provided by the cloud platform. Additionally, we plan to dedicate efforts to implementing dynamic scaling for our application servers, adjusting their capacity in real-time to match the current load conditions.

We are seeking a unified Observability & Management dashboard, a single-pane-of-glass solution, that consolidates metrics from APM, Stack Monitoring, and Logging Analytics capabilities, providing a comprehensive overview in one place.

Describe how your solution was developed to address your top challenges.

The solution was the result of excellent collaboration between cross-functional teams at Big Lots and Oracle with heightened attention to data migration, security, and reusability. We invested time in meticulously architecting the solution and conducting multiple reviews before proceeding with implementation. Remarkably, we successfully completed this migration in under five months, despite the complexity of the tasks involved.

What are the biggest benefits you see with your Observability and Management solution?

We collaborated with Oracle experts to fine-tune the Oracle Observability and Management (O&M) solution to align with our previous monitoring system. We successfully developed dashboards to monitor critical application metrics such as CPU, Memory, and Thread counts, which were like our previous setup. We are eager to see further enhancements in the platform, specifically with the capability to effortlessly incorporate new monitoring checkpoints. We aim to streamline the ability to trace transactions, both at the application and database levels, as this would greatly benefit our support teams.

“Performance is a big goal. We are seeing close to 25 to 30% improvement in performance since we moved from dedicated hardware to OCI.  This is a big positive and makes our site snappier.”                                                     Satish Mukundachar, Director of Applications Development, Big Lots 

What comparisons (pros and cons) can you share from moving your previous monitoring tool to O&M?

Since the OCI O&M was a new tool that was non-existent in the past solution, we worked with an open mind and the fact that O&M Services collected vast amounts of data for analysis. 
The primary benefit of utilizing APM is its integration with the Oracle infrastructure, allowing applications running on OCI to seamlessly access APM capabilities, and ensuring standardized application monitoring.

One opportunity for improvement lies in optimizing our utilization of data within O&M. Currently, the process to access the necessary data is not seamless and often requires multiple working sessions with Oracle to establish new monitoring capabilities. We aspire to have more self-service functionalities that facilitate the transition from legacy to cloud monitoring.

How did Big Lots transform from your previous Monitoring tool to O&M?

During our migration of the Oracle ATG Web Commerce workload to Oracle Cloud, the Oracle team presented various monitoring capabilities offered by Oracle Observability & Management services (O&M). Following a meticulous analysis of our existing system and the O&M features, we had to create custom dashboards to align with the functionalities we were using in our current setup and O&M. 

Our team collaborated closely with Oracle to devise a deployment plan for O&M services, and we successfully implemented the required monitoring services before our production environment went live. We utilized several OCI features in this process:

1.    OCI Stack Monitoring was configured to oversee the infrastructure, while logs from various application components were directed to Logging Analytics for ingestion and dashboard creation.
2.    OCI APM Tracing enabled us to assess the performance of our website and interactions with third-party service providers.
3.    OCI Synthetic Monitoring Scripts were developed to proactively monitor our website from diverse geographic regions.
4.    OCI Alarms were set up across these services to promptly notify our Operations and Support teams of any anomalies.

This has laid foundation to effectively manage the Big Lots website and related services, with the goal of enhancing efficiency in the future. 

What do years two and beyond look like for your solution?

Our objective is to delve deeper into OCI DevOps, streamlining and simplifying our deployment processes. We aim to harness cloud-native features, enabling us to dynamically scale compute resources up or down as needed. This approach will contribute to cost reduction, a key motivator for our transition to the cloud. Additionally, we seek to enhance our monitoring capabilities and establish direct integration between OCI APM and our ticketing systems, ensuring that all alerts generate tickets to our service center.
 

Resources

Kathryn Jennings

Senior Principal Product Manager