Oleg K.
Rating
Language proficiency level
Skills and abilities
Portfolio
-
2000 USD Building and migrating infrastructure in AWS
DevOpsProject Description
For a web project hosted on its own servers (on-premises), the task was to migrate the infrastructure to the cloud with the aim of automating processes, ensuring high availability (HA), and reducing infrastructure maintenance costs.
… What was accomplished:
Design and approval of the new infrastructure: a detailed infrastructure plan was developed and approved with the client.
Cost optimization: the cost of using the cloud platform was calculated and optimized.
Infrastructure as Code: the infrastructure was described using Terraform.
Immutable servers: a strategy for creating immutable servers was implemented using Packer.
Performance testing: load testing was conducted to assess performance and identify bottlenecks.
Step-by-step migration of services: a phased transfer of services was completed.
Compliance with security requirements: the infrastructure was built in accordance with security policies and the requirements of the cloud platform.
Monitoring and alerts: monitoring and a notification system were set up for all critical components.
Backup and recovery: backup and disaster recovery strategies were implemented.
Final testing and documentation: final testing was conducted and all processes were documented.
#Amazon #AWS #terraform #packer #IaC #ansible
-
750 USD Troubleshooting response time issues
DevOpsThe client complained that their website had become slow. My task was to conduct a detailed analysis of the project's infrastructure and services, as well as to propose solutions for optimization or configuration changes to improve the website's loading speed.
What I did:
… I performed a comprehensive analysis of the web server and database configuration.
During the analysis of the MySQL database, the following issues were identified:
The configuration was suboptimal: too small buffers and caches were used for the volume of data. I adjusted the parameters to more effectively utilize the server's resources.
In analyzing slow queries, it was found that some queries significantly increased the load on the server and its disk subsystem. These queries were forwarded to the developers for further analysis and optimization.
I optimized the web server settings and configured the optimal number of Nginx workers to handle requests and efficiently use server resources. A cache in Nginx for static data and part of the dynamic content was also configured.
Another issue was the frequent use of DNS resolution in frequently executed queries, which added delays of several milliseconds. I recommended, where possible, to replace the use of domain names with direct IP address specifications.
There was also a high disk load caused by the web server storing session files on the disk. Redis was installed and configured for session handling.
All these measures allowed for a 30-50% increase in website loading speed and reduced the overall load on the servers, providing the client with resource capacity for further project scaling.
#MySQL/MsSQL/NoSQL #DevOps #Apache/Nginx/PhpFPM
-
1000 USD Setup and implementation of the monitoring stack (Icinga, Prometheus)
DevOpsTask:
For one of the projects, it was necessary to deploy a monitoring stack. The client required monitoring of services and a notification system based on Nagios. Additionally, Prometheus and Grafana were needed to track SLO/SLI and individual services.
… How I implemented it:
Instead of Nagios, I recommended using Icinga due to its more powerful API and user-friendly dashboard. Furthermore, Icinga is better suited for scaling and implementing high availability. The Icinga server was initially configured with scalability in mind. I grouped all services and set up detailed metrics and notifications for the staff.
To collect extended metrics for individual services, Prometheus was used. VictoriaMetrics was employed as the metrics storage due to its high performance and extensive scaling capabilities. A distributed cluster of VictoriaMetrics was deployed for the client.
SLI and alerts for SLO violations were configured for critical services. Grafana was used for convenient visualization of graphs.
#grafana #prometheus #icinga2
Reviews and compliments on completed projects 4
15 September 2021
459 USD
Поддержка веб-проекта Debian/Nginx/FPM/MySQL (интенрент-магазин)
Качественно и быстро. Будем сотрудничать и в будущем.
![]()
8 September 2021
510 USD
Поддержка веб-проекта Debian/Nginx/FPM/MySQL - (300к в день)
Всё отлично, как всегда. Работы выполнены профессионально, будем в будущем сотрудничать.
![]()
14 March 2017
451 USD
Web support for Debian/Nginx/FPM/MySQL - (50-200k visitors per day)
Everything has been done well and we will continue to work with you.
![]()
22 February 2017
56 USD
Toughened Laminated Safety Glass and Nginx+uwsgi and gitlab
We planned for two days, the work was done in a couple of hours. Everything according to requirements
![]()
| Response review
Activity
| Latest proposals 10 | Budget | Added | Deadlines | Proposal | |
|---|---|---|---|---|---|
|
System Administrator / DevOps (hosting, Linux, high-load)
451 USD
|
|||||
|
Working with the server
23 USD
|
|||||
|
VPS server setup
23 USD
|
|||||
|
Need a devops quickly
23 USD
|
|||||
|
Установка ПО, настройка чистого сервера под сервер раздачи и настройка failover, через CloudFlare
338 USD
|
|||||
|
Restoration of MySQL replication + integration with Power BI
338 USD
|
|||||
|
Поддержка веб-проекта Debian/Nginx/FPM/MySQL (интенрент-магазин)
459 USD
|
|||||
|
Поддержка веб-проекта Debian/Nginx/FPM/MySQL - (300к в день)
510 USD
|
|||||
|
Web support for Debian/Nginx/FPM/MySQL - (50-200k visitors per day)
451 USD
|
|||||
|
Cleaning the site from scripts
23 USD
|