Matoffo

Matoffo Solution

Since most of Client's services have a similar structure, we decided to start by splitting each Laravel PHP service into separate components (web server, fmp server, queue, cron job). Next, we dockerized each of these components. Kubernetes Cluster was created to manage docker containers.

As a core of this solution, our development team used AWS EKS because it provides a fault tolerance/high availability environment with multiple beneficial features like the ability to use:

• Scalable solutions using HPA/Replicas/Specified resources per Service;

• Karpenter as Autoscaler to AWS EKS Cluster (can determine which node should be raised with a specific type);

• Managing secrets using HashiCorp Vault (injected as a sidecar to Kubernetes pods);

• Automation deploy using GitOps solution (ArgoCD as a tool to implement this approach);

• Full automated CD with automatically created DNS records (using ExternalDNS addon for EKS);

• Clear monitoring and logging using Prometheus and Grafana (with own dashboards/alert manager to get some notifications in Slack);

• Ability to create new environments from scratch using the IaC approach (Terraform for infrastructure/Kustomize for applications).

Last but not least, we also built a VPN server based on OpenVPN to enable completely private access to the AWS environment and all services located in Private Subnets and not exposed to the world.

Additionally, we redesigned the GitLab CI part to maintain a fully automated build and deploy services without interruptions so that the developer just pushes code, while CI and CD completely cover all the rest.

The development team included three DevOps engineers responsible for different processes, including Kubernetes configuration, CI/CD setup, frontend development, and allied support. In general, it took us about a month to deliver a ready-to-use AWS solution.

Please see more details here -

https://matoffo.com/cases/aws-native-kubernetes-solution/

The client had high costs for development, deployment, and, most importantly, operation of the data platform including Data Lake, Ingestion, and ML Pipelines. The pipelines were mostly running in EC2 instances, which led to the increased cost of operations and required a significant amount of time to deploy and test the pipelines in lower environments.

The solution allowed to significantly reduce the cost and time spent on development and deployment. It was achieved by introducing infrastructure as code that quickly and seamlessly deploys infrastructure on multiple environments (Dev, Stage, Production). Besides, during the implementation, the team developed the set Lambda and Step function which triggers data processing by the event that excludes cases where EC2 instances are idling and created cost-efficient Data Lakes.

Full version of Case Study - https://matoffo.com/cases/serverless-data-ml-pipelines/

Matoffo Solution

There was created a new AWS environment based on microservice architecture. The docker image was developed according to the latest best practices like multi-stage building, minimizing the number of layers, decoupling applications, etc. Backed services were deployed using AWS Fargate. For the front-end, the web application was configured as a CDN service using AWS CloudFront. To automate deployment and release processes, we created GitLab CI pipelines.

All secrets for the deployment process are stored securely in GitLab CI while all infrastructure is deployed by Terraform. State files for Terraform are stored in AWS S3 so they can be easily accessible from local and CI environments. For solving requests to deploy with zero downtime and rollback, we have decided to use AWS CodeDeploy. This solution gives the possibility to use Blue/Green deployment and an easy way to rollback to the previous version.

To complete this solution, we used the following tech stack: Amazon S3, RDS, Terraform, Gitlab CI, Application Load Balancer, Amazon ECS, CloudFront, Docker, ElastiCache, AWS CodeDeploy. The development team included three DevOps engineers (provided by Matoffo) and one technical director (provided by the client). In general, it took us about a month to deliver a ready-to-use AWS solution.

Please see more details here - https://matoffo.com/cases/a-highly-available-microservice-architecture/

The client experienced the need to automate the deployment process of microservices and simplify the release process. One of the requirements was to migrate Docker containers to AWS ECS and move existing infrastructure to Terraform. 

With the delivered solution, the client has got a highly available and resilient to failures microservice architecture that uses AWS ECS and a fully managed and automated deployment process using Jenkins CI. Besides, the infrastructure has been migrated to Terraform which allows to manage and add new resources much easier and more efficiently.

More details here -

https://matoffo.com/cases/infrastructure-devops-services-for-fintech-product/

Initially, the client's development team performed continuous integration and continuous deployment processes manually. Consequently, it was time- and effort-consuming to build cross-platform mobile applications. Besides, they used the legacy tools stack – that doesn't have built-in mobile-specific continuous integration and deployment capabilities.

The introduced solution significantly improved the time to release, and also enhanced the mobile application development itself by automating continuous integration and continuous deployment process that positively impacted business KPIs

Full version of Case Study - https://matoffo.com/cases/cicd-for-cross-platform-mobile-development/ 

The client faced the need to automate multi-environment continuous integration and deployment process of microservices, move the front-end part to CDN as well as deploy multi-environment infrastructure as a code.

With the delivered solution, the client has got a highly available and resilient to failures microservice architecture powered by AWS ECS with CloudFront and a fully managed and automated deployment process powered by GitLab CI.

Full version of Case Study - https://matoffo.com/cases/iaccicd-cloud-devops-services-for-real-estate-product/

The client needed to strengthen collaboration within the data scientists team by moving research capabilities into cloud workloads. Besides, they wanted to automate and unify the deployment process of AWS resources. That, in turn, would decrease the time and effort required for a team of data scientists to build and test their models.

The developed solution has allowed the client to use Terraform code and GitLab pipelines instead of manual AWS infrastructure and lambdas deployment. The data scientists team can now deploy AWS resources and test models in cloud workloads faster and more efficiently, which means they can focus on key scientific objectives rather than cloud engineering.

Full version of Case Study - https://matoffo.com/cases/iac-data-science-infrastructure/

The client had a manual multi-environment infrastructure deployment, аnd no continuous integration & continuous deployment automation. 
As a result, implemented solution significantly increased the speed of product development, release and delivery cadence of new application versions and, finally, allowed to fully focus the development team capacity on product development rather than on doing manual CI/CD activities. 

Full version of Case Study - 
https://matoffo.com/cases/iaccicd-cloud-devops-services-for-cloud-native-product/

CLIENT BACKGROUND

The client is the one of leaders in telecommunication services domain by delivering the unique value proposition in targeted business niche. To deliver the services utilizes data center infrastructure with comprehensive hardware and software solutions.

CHALLENGES

The client experienced the need to migrate as much of on-premises IT infrastructure resources which are being running on an outdated VMware ESXi cluster (dedicated physical boxes) as possible due to shutdown of a classic data center in the current physical location. The environment included a plethora of different services to migrate: AD DS, DFS, DHCP, File Servers, old custom 2-tier applications, BizTalk application cluster, IIS, TFS server, SQL Servers, etc.

SOLUTION

Has been proposed to use the built-in AWS capabilities and tools. Hence, the choice was made in favour of AWS server migration service. Primarily, most of the services were migrated using lift and shift strategy. Once the VM images were uploaded to the cloud, Terraform was used to create the needed resources and perform initial configuration. Minor architectial changes were made to DFS and DHCP. SQL Servers were migrated to AWS RDS PaaS offering. 

TECHNOLOGIES

HashiCorp Terraform, AWS RDS, AWS EC2, AWS CloudTrail, AWS Migration Service, AWS Database Migration Service.

BENEFITS

All services are smoothly running in the cloud and as a result the client got a better performance with even lower specs due to smaller network latency and better rightsizing. Switching to RDS allowed to release the burden of OPS tasks. From security perspective all migrated services are aligned with cloud environment, leveraging storage encryption, improved network security. Additionally, highlighting that improved downtime KPI, business agility and significantly decreased capital expenditure. 

CLIENT BACKGROUND

The client is one of the biggest banks in the Eastern Europe. Utilizes comprehensive solution which includes multi-purpose payment system that combines cashier, accountant, clerk, and administrator workplaces. Additionally the application service is an engine for self-service kiosk machines. The system processes millions of payment transactions per month, aggregates and keeps track of a complete bank day workflow, the number of payment during month significantly vary, in the beginning of a month the load may be 5x-8x larger than daily average.

•  4 large Microsoft Hyper-V application services must be migrated to AWS for automatic scale in/out

• 4-node Microsoft SQL Server DB cluster must stay in the client's datacenter but be replicated to AWS

• 2-node DWH and Reporting Service machine must be migrated to AWS

CHALLENGES

The goal was to partially migrate on-premise infrastructure. Application services must be hosted in the cloud for cost reduction and scaling. The database cluster must be replicated in the cloud to reduce costs of disaster recovery infrastructure and reduce downtime, but the write nodes should stay in the local data center.

• Diverse infrastructure, complex security rules, thousands clients over LAN

• Legacy Application (Borland C++). Possible instance failures, no automatic disaster recovery.

• Complex database with heavy usage of triggers, transactions and stored procedures.

• Regular spikes in the workload

SOLUTION

AWS Migration Service helped to migrate application services. The VM incremental backups allowed prolonged iterative step-by-step migration with an excessive testing phase. The EC2 auto scaling groups enabled cost-effective automated scale in/out approach. CloudEndure Disaster Recovery provided easy way to establish bridge between on-premise SQL Server cluster and backup replicas. VPC subnets, security groups and ACLs allowed quick and secure infrastructure configuration for further operational usage.

TECHNOLOGIES

AMS, CloudEndure DR, EC2 + Auto Scaling, Cloudformation, VPC, ​Migration Service.

BENEFITS

As a result, the solution helped the client to significantly reduce infrastructure costs and strengthen customer satisfaction via improved business continuity approach. Auto scaling, improved backup strategy and disaster recovery were the key differentiators that helped the client with a decision on further cloud migration. As a consequence, though the initial phase was critical for the company's path to digital transformation, but with the help of AWS migration services the client ensured that the cloud migration is right path and trusted Matoffo and AWS with further cloud strategy development. Migration features like incremental replication allowed to test the new infrastructure and smooth iterative transition before the final cutover, which release the overall migration stress.

The purpose of the solution to gain maximum benefits from standing desks, majority of users doesn't use them correctly (e.g. stands or sits a lot without breaks) which make the usage unhealthy. To solve this issue have been developed solution which included mobile, backend and hardware compoments.

Technologies: Node.js, React.js, Nem Blockchain, Android.

The main purpose of the solution is to support comprehensive business processes during initial coin offering. The primary focus was on token sale flow, it supports cryptocurrency transactions as well as fiat currency transactions. The token sale flow takes into account many factors that increase conversion. Apart from that, the solution includes other significant functionality, for instance, which intended to support KYC criteria and requirements, other functionality.

 Technologies: React.js, Node.js, Smart Contracts, Blockchain.

Supply chain solution for transportation goods which are sensitive to temperature, humidity, and other parameters during transportation. The main goal of the solution to make transparent shipment parameters for participants and ensure that data is immutable. The application runs on top of Hyperledger Blockchain which is intended to provide transparency for all parties and provides data storing immutability. 

Technologies: Hyperledger, AWS, Go, ReactJS.