Code Impact

Code Impact

By Sanket MakhijaEducation
Download on the App Store

All Code Impact episodes

  • The Practical Guide to GraphQL Adoption in PayPal

    GraphQL has generated a lot of buzz, but what does it take to successfully adopt it within your organization? This podcast cuts through the hype and provides a practical roadmap for GraphQL implementation. Drawing on PayPal's experiences, we'll explore the key factors to consider, from assessing your company's needs and setting realistic expectations to building a solid foundation, scaling knowledge, and establishing design standards. We'll also address common challenges, such as performance optimization, error handling, authentication, and the importance of investing in the GraphQL community.

    9 min
  • Understanding the Gitflow Workflow

    The Gitflow workflow is a branching model for Git that can be used to manage the development of software projects. In this episode, we will discuss the different branches in the Gitflow workflow and how they are used. We will also provide an example of how to use the Gitflow workflow to develop a new feature.


    Reference: https://www.atlassian.com/git/tutorials/comparing-workflows/gitflow-workflow

    8 min
  • Geospatial Analytics at Scale with Presto

    Geospatial analytics is becoming increasingly important as companies seek to gain insights from location data. This podcast will discuss how Presto can be used to perform geospatial queries efficiently. We will also discuss the QuadTree data structure, which is used to index geospatial data in Presto.

    Reference: https://scontent.fblr1-4.fna.fbcdn.net/v/t39.8562-6/280858273_398229132019114_3101674654302651_n.pdf?_nc_cat=108&ccb=1-7&_nc_sid=e280be&_nc_ohc=_DdBGwQNZnIQ7kNvgGlIBGq&_nc_zt=14&_nc_ht=scontent.fblr1-4.fna&_nc_gid=AOTWRBja0lnMACuRisrKSQr&oh=00_AYA6MfDxtmeCQTL_-bQrVSEPZbBBA8XqiDGs0TvV1m9nMw&oe=6715C2B6

    19 min
  • From Likes to Less Servers: Facebook's Journey to MyRocks

    This podcast explores Facebook's transition from using the InnoDB storage engine to MyRocks, an LSM-tree-based storage engine built on RocksDB, for its massive User Database (UDB). The podcast will examine the challenges Facebook faced with InnoDB, such as index fragmentation and write amplification, and how MyRocks helped overcome these obstacles. It will discuss the key features and optimisations of MyRocks, including prefix bloom filters, tombstone management and bulk loading, and how these innovations led to a 62.3% reduction in instance size and fewer database servers.


    Reference: Research paper "MyRocks: LSM-Tree Database Storage Engine Serving Facebook's Social Graph"

    16 min
  • Architecture of a Database System: Tech Deep Dive

    This episode explores the intricate architecture of Database Management Systems (DBMSs), those mission-critical software systems that power our digital world. From the earliest online server systems to modern-day giants, we trace the evolution of DBMS design, examining key components such as process models, parallel architectures, storage systems, transaction systems, query processors, and optimizers. Get ready for an insightful journey into the heart of database systems!


    Reference: https://dsf.berkeley.edu/papers/fntdb07-architecture.pdf

    19 min
  • What Every Programmer Should Know About Memory
    • Join us as we explore the intricate world of CPU and memory architecture, uncovering the hidden performance bottlenecks that plague software developers.
      • We'll break down complex topics like cache behaviour, virtual memory, NUMA systems, and more, empowering you to write faster and more efficient code.
      • From understanding the impact of cache lines and TLB misses to leveraging advanced techniques like prefetching and helper threads, we'll equip you with the knowledge to optimise your code for maximum performance.
      • Whether you're a seasoned developer or just starting out, this podcast will provide valuable insights and practical tips to elevate your coding skills and create high-performing applications.

      • Reference: https://people.freebsd.org/~lstewart/articles/cpumemory.pdf

        11 min
      • Beyond Zstandard: Unexpected Wins in Discord's Performance Journey

        Discord's pursuit of a more efficient platform led them down a path filled with unexpected discoveries. This episode goes beyond the implementation of zstandard compression to explore the surprising optimisation of passive sessions. We'll analyse how a keen eye for detail and a data-driven approach uncovered hidden opportunities for improvement, resulting in substantial bandwidth savings


        Reference: https://discord.com/blog/how-discord-reduced-websocket-traffic-by-40-percent

        10 min
      • From Monolith to Microservices: A Razorpay Tech Deep Dive

        Get ready for a technical deep dive into Razorpay's evolution from a monolith to a microservices architecture.

        Our guest, Arjun, a payments platform engineer at Razorpay, walks us through the pivotal decisions and engineering feats that enabled Razorpay to scale its systems and handle the massive transaction volumes of India's booming digital payments landscape.

        We'll explore the challenges of splitting databases, the importance of the outboxer pattern and CDC pipelines for maintaining data consistency, and the intricate testing strategies used to guarantee system reliability.

        This episode offers invaluable insights for engineers grappling with similar challenges in their own microservices journeys.


        Reference: https://youtu.be/yqkyq8TPWbg

        13 min
      • ● Apache Kafka, Flink and Pinot: Open Source Powering Uber's Real-Time Data Stack

        Building and scaling a real-time data infrastructure is a complex undertaking, fraught with challenges and valuable lessons. This episode takes a deep dive into Uber's journey, exploring the hurdles they encountered while managing petabytes of real-time data. We'll discuss the need for data consistency, availability, and freshness, the complexities of handling diverse use cases and user groups, and the constant need for system evolution. Tune in to learn from Uber's experiences and gain insights into building robust and scalable real-time data infrastructures.


        References: https://arxiv.org/pdf/2104.00087

        12 min
      • Instant Purge: How Cloudflare Makes Content Disappear in the Blink of an Eye

        Imagine a world where online content updates instantly, everywhere. This podcast explores Cloudflare's groundbreaking "Instant Purge" system, which removes outdated content from its cache in under 150 milliseconds. Discover the innovative technologies and strategies behind this feat, and how they impact the internet experience for users worldwide.


        Reference: https://blog.cloudflare.com/instant-purge/

        11 min