In modern microservice architectures, selecting the right protocol for interaction between microservices plays a crucial role in the overall performance and reliability of the system. In my presentation, I will share our migration experience from the traditional HTTP API to gRPC, including the motivations behind this decision and the challenges encountered during the migration process.
We will explore the advantages and limitations that gRPC brought to our system and the results of an experiment where we compared the performance of the protocols.
Through real examples, we will check some use cases where migration to gRPC is a great solution and discuss scenarios where this approach might be less effective.
"Key considerations in implementing a distributed message-sending system usin...Fwdays
油
Designing and implementing a scalable and reliable message-sending service may seem like a challenging and time-consuming task. However, let's explore some key points of implementation in .NET that will help us achieve the desired level of quality and avoid unexpected obstacles.
To accomplish this, we will:
Explore of some features of the .NET Confluent Kafka driver.
Examine real-life use cases of utilizing .NET channels as an InProc Pub/Sub mechanism to enhance application performance.
Discuss the usage of Minimal API and understand its limitations.
Compare gRPC streaming with HTTP and determine which option is more suitable for our specific scenario.
"Request Lifecycle at Prom.ua", Vitaliy KharytonskiyFwdays
油
A detailed journey of a single user request through prom.ua infrastructure including hardware, networks, services, and databases.
This talk will feature an explanation of our architecture and reasons behind its current state and external and internal challenges which arise in a high load project built completely on premise.
We at MacPaw practice the approach of service teams. And as one of such teams, the responsibility for delivering software across various environments, from testing to production, falls on our shoulders. In this presentation, I will explain how we are trying to standardize our approach to software delivery in environments with diverse tech stacks and development approaches. The presentation will cover methods such as GitOps, dynamic environments, and event-based software delivery.
Pro Magento MeetUp #9 - Deep dive into Magento2 queuing functionality.pdfPro Magento Community
油
Topic: Deep dive into Magento2 message queues
Speaker - Yevhenii Trishyn, Magento Senior Full-Stack Developer, IT Delight
What we discussed:
- What are message brokers, where and how are they used
- #AMQP protocol and its description
- RabbitMQ and its description
- How Magento2 works with RabbitMQ
- Where this functionality is used in core modules
- How to add queue support to your module
Recording of the MeetUp - https://youtu.be/XsbIOLwHnaE
Join our community in Telegram:
chat for communication - https://t.me/+lfA4OJHlSWozYWY6
news channel, articles and announcements of Pro Magento events https://t.me/promagento_channel
"Simplifying the Complex: Effective Management of Large-Scale PHP Projects", ...Fwdays
油
Developing and managing large projects can become a challenge for developers. This presentation will show how to overcome these challenges, using the example of one of the largest open-source PHP projects. We will discuss approaches to code organization, coding standards, effective tools, as well as the role of code review and testing. We won't forget about design patterns either.
Oleksandr Brychuk "UniSender architecture. Growth from 100kk to 1.5kkk letter...Fwdays
油
Architecture in 2013 comes from scratch, so is there any hope for the future? Business is primarily about money, but what if the balance between technical improvements and a beautiful look is not maintained.
"Emergency 2015" - the limit after which you need to make drastic changes in the technical component. Carte blanche from business and a rough idea of where to start the transformation.
Why did you choose to go through refactoring? Why did you decide to split the monolith into microservices in 2015, when the hype was just emerging, instead of SOA and monolith? How did you choose where to start? How AWS S3 defeated Ceph and helped save the nerves and funds of DevOps? What nodes of the system have provided us with the opportunity to grow 10-15 times in 5 years without spending much more money on vertical scaling? Stable 1.5 billion letters in 2020.
"Spin-up pgbouncer for fun and profit", Vitaliy KharytonskiyFwdays
油
In this presentation I will try to analyze the experience of using pgbouncer and other database proxies at prom.ua. What profit does it bring in organizational sense, what new flexibility it gives to the infrastructure and everyday maintenance and scaling, which additional enchancements we managed to implement and test, what worked and what did not.
"Building Trust: Strengthening Your Software Supply Chain Security", Serhii V...Fwdays
油
The talk focuses on developing and integrating automation tools to enhance Supply Chain security. It addresses reproducible security practices with tools like Renovate and Wiz, as well as GitLab and JFrog Artifactory, to enforce consistent security scans seamlessly within existing workflows.
We will cover centralized artifact management for improved oversight and consistency. Furthermore, we will discuss the seamless integration of security scans into deployment tooling, featuring automatic deployment blocks for vulnerabilities and a controlled override option for flexibility.
The talk also examines tactics to keep the source code secure and up to date. We will explore the integration of runtime monitoring systems with detection capabilities and SLAs to manage and resolve issues on time.
More Related Content
Similar to "gRPC-based microservice communication: pros and cons based on our experience", Ihor Zakutynskyi (20)
We at MacPaw practice the approach of service teams. And as one of such teams, the responsibility for delivering software across various environments, from testing to production, falls on our shoulders. In this presentation, I will explain how we are trying to standardize our approach to software delivery in environments with diverse tech stacks and development approaches. The presentation will cover methods such as GitOps, dynamic environments, and event-based software delivery.
Pro Magento MeetUp #9 - Deep dive into Magento2 queuing functionality.pdfPro Magento Community
油
Topic: Deep dive into Magento2 message queues
Speaker - Yevhenii Trishyn, Magento Senior Full-Stack Developer, IT Delight
What we discussed:
- What are message brokers, where and how are they used
- #AMQP protocol and its description
- RabbitMQ and its description
- How Magento2 works with RabbitMQ
- Where this functionality is used in core modules
- How to add queue support to your module
Recording of the MeetUp - https://youtu.be/XsbIOLwHnaE
Join our community in Telegram:
chat for communication - https://t.me/+lfA4OJHlSWozYWY6
news channel, articles and announcements of Pro Magento events https://t.me/promagento_channel
"Simplifying the Complex: Effective Management of Large-Scale PHP Projects", ...Fwdays
油
Developing and managing large projects can become a challenge for developers. This presentation will show how to overcome these challenges, using the example of one of the largest open-source PHP projects. We will discuss approaches to code organization, coding standards, effective tools, as well as the role of code review and testing. We won't forget about design patterns either.
Oleksandr Brychuk "UniSender architecture. Growth from 100kk to 1.5kkk letter...Fwdays
油
Architecture in 2013 comes from scratch, so is there any hope for the future? Business is primarily about money, but what if the balance between technical improvements and a beautiful look is not maintained.
"Emergency 2015" - the limit after which you need to make drastic changes in the technical component. Carte blanche from business and a rough idea of where to start the transformation.
Why did you choose to go through refactoring? Why did you decide to split the monolith into microservices in 2015, when the hype was just emerging, instead of SOA and monolith? How did you choose where to start? How AWS S3 defeated Ceph and helped save the nerves and funds of DevOps? What nodes of the system have provided us with the opportunity to grow 10-15 times in 5 years without spending much more money on vertical scaling? Stable 1.5 billion letters in 2020.
"Spin-up pgbouncer for fun and profit", Vitaliy KharytonskiyFwdays
油
In this presentation I will try to analyze the experience of using pgbouncer and other database proxies at prom.ua. What profit does it bring in organizational sense, what new flexibility it gives to the infrastructure and everyday maintenance and scaling, which additional enchancements we managed to implement and test, what worked and what did not.
"Building Trust: Strengthening Your Software Supply Chain Security", Serhii V...Fwdays
油
The talk focuses on developing and integrating automation tools to enhance Supply Chain security. It addresses reproducible security practices with tools like Renovate and Wiz, as well as GitLab and JFrog Artifactory, to enforce consistent security scans seamlessly within existing workflows.
We will cover centralized artifact management for improved oversight and consistency. Furthermore, we will discuss the seamless integration of security scans into deployment tooling, featuring automatic deployment blocks for vulnerabilities and a controlled override option for flexibility.
The talk also examines tactics to keep the source code secure and up to date. We will explore the integration of runtime monitoring systems with detection capabilities and SLAs to manage and resolve issues on time.
"10 Pitfalls of a Platform Team", Yura RochniakFwdays
油
There are many obstacles and pitfalls on the path towards operational zen. Many routes could lead to dead ends and many detours could end up being loops.
In this semi-comedy talk I share some examples of how processes fail for engineering teams that I observed through my career, using two pillars of the Internet culture - memes and numbered lists.
"Reality of Managing 100+ Managed RDS Postgres Databases", Mykyta HlushakFwdays
油
During the lecture, Mykyta will share the story of Solidgate's journey in building a high-performing and reliable fintech company, striving for 99.999% uptime on AWSs SaaS platform. He'll uncover numerous caveats in doing things right without full system access, addressing product requirements, and staying up to date.
"Zero-sales lost Incident Management", Igor DrozdFwdays
油
In my presentation, I will discuss the key aspects of the Zero Sales Lost strategy, focusing on incident management in the context of E-commerce development and the "Silpo" ecosystem. I will share our experience in building an incident management process, the specifics of monitoring and alerting system architecture, and how we scaled these processes across 19 products and 41 teams.
Additionally, we will explore the role of DevOps culture and team responsibility in product development. Using real case studies, we will analyze whether critical failures can be prevented and how to achieve that effectively.
"Turning Kubernetes into a full-fledged private cloud", Volodymyr TsapFwdays
油
At the conference, I will present an approach to extending the capabilities of Kubernetes, transforming it into a fully-fledged private Cloud with support for virtualization, isolated networks, and multi-organization authentication.
The goal of the presentation is to demonstrate that Kubernetes is not just a platform for containers, but a fully-featured private cloud that can be easily adapted to business needs.
"Kubernetes operators. How we migrated Release Management to controllers", De...Fwdays
油
Kubernetes Controllers and Operators are a trending topic in conferences, interviews, and production today. I will share the story of the evolution of our Promotion (Release) system, from simple Kubernetes API REST calls to Informers and Controllers, based on my own experience. This story is particularly interesting because it serves as a great case for personal growth for you as well as for your DevOps/SRE team. It touches on Kubernetes architecture details, Networking, GitOps, IaC, Caching, development patterns, and Golang data structures. Even if you have no development experience (as is the case for most of our team), I will share how a Cursor AI assistant became yet another though virtual engineer on our team.
Bonus: 10 years of Kubernetes & trends KubeCon24 North America.
"How AI infrastructure differs from common infrastructure", Vsevolod PolyakovFwdays
油
For the past three years, I have worked at LetsEnhance/Claid, an AI startup that enhances image quality and creates new backgrounds for customer products. While the infrastructure is always the same, it varies: large models, significant latency, and unreliable autoscaling. In this speech, I will discuss our perspective on AI infrastructure and our ideas for its improvement.
The topic is for mid to senior-level engineers who are either already in AI or interested in understanding how it differs from traditional infrastructure.
"DevOps culture and digital transformation process of Temabit Fozzy Group", O...Fwdays
油
The transformation of a large organization is a marathon where every step requires coordinated teamwork, creative solutions, and a willingness to embrace mistakes.
We will discuss:
- How to properly plan global changes in a company with over 25,000 clients.
- Why communication between technical teams and the business is the key to success.
- SRE, R&D, and Architecture as drivers of transformation.
- The role of FinOps in sustainable development.
- Real-life cases, mistakes, and lessons learned.
This will be a story of ups and downs that helped us turn challenges into opportunities.
What we've learned by implementing a domestic microservices system for 5 year...Fwdays
油
I'll share what problems our team faced in the process of developing and maintaining a microservice architecture software system for 5 years, what solutions we applied and how it affected us.
In the talk we'll cover: domain modeling, assets topology, value of standards and soft skills.
"Transition from Bench to Reserve: how to effectively manage and motivate peo...Fwdays
油
Sergiy will tell you how to change your perception of the company's reserve. Don't be afraid to have it and don't go broke on it. He will share how he personally works with the reserve and how to form it correctly and incorrectly. What mistakes are most common in companies of any size. He will tell how to motivate people and make them understand that the reserve is not a way out of the company, but rather the opposite
"Money and Startups: How to Get $200,000+ in Equite Free Funding for Startup ...Fwdays
油
The differences between grants and investments: which is better for your startup?
How to raise money without selling equity, using the Dots Platform experience, including:
How we recieved $50k from the Ukrainian Startup Fund.
How we recieved up to $100k from Google for Startups.
How to leverage AWS and Google credits to reduce costs, how we received credits for over 3+ years for our infrastructure, and how we made the most of them.
Our lessons in scaling to international markets: the experience of entering the Irish and US markets, and when investment doesnt solve all the problems.
"Slow but Steady vs Fast and Furious: The Dilemma of Modern Development", Kyr...Fwdays
油
Let's talk about one of the most painful topics finding the balance between the speed of developing new features, technical optimization, and managing technical debt.
Using the example of our startup's development, I'll share how we tried to rush forward, when we realized something was going wrong, what mistakes we made, and what it led to. I'll discuss how we solved the accumulated problems and what we got from it.
Finally, we'll discuss what could have been done differently from the very beginning and how to search for that elusive balance in general.
As a CTO, I often faced the temptation to implement new technologies. This technological roulette sometimes led to success, sometimes not, but it always provided valuable experience. Our product continues to evolve, which indicates more wins than losses. Today, Ill share the lessons we learned from this game and explain how to distinguish valuable innovations from risky experiments.
"Optimization of Retrieval-Augmented Generation (RAG) for eCommerce: Personal...Fwdays
油
A practical case study exploring how fine-tuning RAG reduces the risk of AI hallucinations and improves response accuracy by combining knowledge from a database with dynamic API requests
"Secrets of Scaling a Product from 2 to 30 Teams", Igor Drozd.pptxFwdays
油
In this presentation, Igor will unveil the behind-the-scenes journey of developing Silpo's E-commerce and Ecosystem. You will learn how they transformed their product and business, overcoming todays challenges while continuing to grow actively. He will share the key decisions and strategies that enabled them to scale from 2 to 30 teams in less than three years, along with the difficulties they faced along the way. This is a story about challenges, growth, and the power of teamwork that allowed them not only to endure but to create something greater, even in the midst of a full-scale war.
"Ukrainian cloud and AWS/Azure/GCP: Smart vs Porsche. Is the alternative real...Fwdays
油
This talk will address the issue of choosing between global hyperscalers like AWS and local cloud solutions for businesses and developers in Ukraine. We will consider what to do if resources are limited or AWS cloud services become too expensive due to unpredictable fees. We will discuss the pros and cons of building your own cloud, the risks of this solution, and how to ensure its reliable support.
The presentation will focus on Ukrainian cloud solutions: from classic IaaS services to modern Kubernetes-based solutions for developers and specialized platforms for AI/ML projects. We will also consider how such solutions are priced and what approaches to user support are the most effective.
This presentation will help businesses and developers find the best way to use cloud technologies in the Ukrainian context.
"5 years at RozetkaPay in 30 minutes: get it all done", Oleksandr Tarasenko.pdfFwdays
油
CTO's experience in a fintech company from the beginning of the project to the top position in the market. With all the regulations and dynamics in Ukrainian realities
"Reducing Incident Impact: Practical Case Studies", Dmytro Dziubenko.pptxFwdays
油
This talk will cover real cases of incidents in a B2B environment and strategies to reduce their scale. Youll learn practical approaches to building resilient and scalable architecture that helps minimize business risks.
23. 亠亰仍舒亳 亠从仗亠亳仄亠仆
HTTP Post request gRPC procedure call
100k async requests, sec 32.342 4.370
Requests per second 2147 26102
CPU usage ms/req 491 ms / per request 211 ms / per request