Open Source-(Sys|Dev)ops Engineer
- Name: Dominique CARREL
- Nationality: French
- Age: 41 (DoB:05/02/79)
- Experience: +15 years
- Email: dpac_at_citoyx.com
- Phone: +33 (0) 646221297
- Web: https://www.citoyx.com
- Stackoverflow: http://unix.stackexchange.com/users/128153/netmonk
- github: https://www.github.com/netmonk/
I have wide knowledge from theorical computing to high avaibility industrial platforms, from computer security to kernel internals.
The past few years, my main passion is about system performance, on how to achieve the best results from a defined starting point.
I’m looking forward for any kinds of positions worldwide, as contractor, consultant, remote worker, anywhere system performance are concerned, helping to achieve the most with less.
I felt in love with the Open Source movement at the age of 17 (my first Slackware, and my first FreeBSD), and have remained vigilent in my open source education.
I like challenging environment, where every issue, every request, every project, are a new journey to grow my knowledge, while delivering the best results to customers.
- Linux (Debian,Redhat, Archlinux, Centos, Ubuntu)
- Timekeeper, Sfptpd, Apache, nginx, bind, squid, jabber, proftpd, haproxy,nodejs,
- clustering, Postfix, qmail, google app gmail for small business
- TCP/IP, Kernel Bypass technologies, firewalling, loadbalancing, vpn (ipsec/openvpn), stunnel,
- Zyxel appliance, HTTP, PTP, FTP, MAIL, POP, SNMP, DHCP, Routing, Switching, vlan.
- Git, Ansible, gcc, gdb, perf, jenkins, groovy, and all standard unix commands
- C, ASN.1, ShellScripting (sh/bash), AWK, PYTHON, ASM x86, NODE.JS
- PAYMENT PROTOCOL (CBCOM, CB2A, APACS, AMERICAN EXPRESS, EPAS), PAYMENT INDUSTRY AND APPLICATION
- TRADING PLATFORM, HIGH FREQUENCY TRADING, HIGH PERFORMANCE COMPUTING
- CIARA HIGH PERFORMANCE SERVERS, HP HIGH PERFORMANCE SERVERS
- MELLANOX TECHNOLOGY 10G NIC Cards and McoreRT
- SOLARFLARE 10G NIC Cards and openonload kernel bypass technologies.
- 2005 Brainbench Unix system administration certification
- 2013 ITIL Foundation certification
[02/2019 - 02/2020] Automation Technical Leader (Credit Agricole/CAPS)
Designing and implementing Continuous Integration Platform as a POC for new Emitter Authorization Platform. CAPS started a new project to change their main payement authorization platform. I designed the full infrastructure for automated deployment from qualification stage to production stage.
In a very challenging environment, strongly restricted by design (PCI/LPM/OIV) the goal was to introduce new ways to deliver applications, deploy them, configure them and testing from receiving a binary package from software company up to putting it into production. With automation at heart, reducing heavy human work, accelerating delivery speed, improving test quality and easing daily maintenance by normalizing environments. At first this was a specific Proof of Concept (POC) dedicated to this project, in the final stage to handover all the experiences and knowledge acquired up to this point, to a wider core project of creating better tooling/process shared across hundreds of projects
Environment: Redhat, Git/GITLAB, Jenkins, Ansible.
Notes: Developed a new feature in the Mysql module of Ansible, which was accepted and merged in the comming release 2.10 of Ansible.
[10/2017 - End of 2018] Infra engineer, SILCA (Credit Agricole)
- Managing the Openstack Cloud platforms and Ceph cluster of the innovation department of SILCA.
- Automating everything using Ansible development and Jenkins deployment
- Troubleshooting VM performance Issue, and incident management.
- Testing and deploying new tools to manage and monitor thoses platforms (In charge to test and deploy Prometheus and connect it with grafana)
- Cattle (Redhat) philosophy of managing huge numbers of servers and clusters
Environment: Centos, Ceph, Openstack, Git, Ansible, Prometheus, Sensu, Awk/Bash scripting, HP servers, bigdata/devops. Volume: 150 Openstack Compute node, 10 petabytes Ceph cluster.
[07/2017 - 10/2017] Performance engineer, EDELIA
Evaluating new releases performance, and validating no regression for the Equilibre Portal of EDELIA.
Troubleshooting key component performance regression under Windows server
Environment: Dynatrace, Neoload, Redhat Linux, Git, VMware
[04/2015 - 06/2017] System engineer, and R-D at SOCIETE GENERALE
Managing Linux servers of the EMM low latency trading platform (formely called High Frequency Trading).
Due to choice of trading strategies, I had to solve the delicate problem of being very fast and reliable.
In short “being the fastest as possible all day long !!!”
- Managing over 12 colocations all over Europe (around 150 highend servers mostly doing Real Time Market Access, spread in 7 differents countries)
- Keeping the platform in production state, and enforcing the compliance with Societe Generale standards.
- Break nothing !!! (Neither incident, nor service interruption since 1st of January 2016)
- Managing Bangalore team which performs the level 1/2 on this plateform (heavy delegation work, technical writing, procedure writing)
- Troubleshooting any kind of issues (network, system, performance, hardware failure)
- Communicating with the dedicated network team and application support team to solve issues, and increase performance.
- Managing market data recorder and storage (>50TB of data) for backtesting, and simulations
- Scientific approach of measuring the impact of performance from tiniest change
- Deep investigations on performance variations
- Testing various hardwares for improving the tick to order of the platform
- Testing third parties technological solutions (network card/kernel bypass solution, high performance servers, cpu…) to improve latency
- Communicating with third parties R&D and support
- Kernel performance tuning and intense technological monitor on new solutions
- Assisting developpers and traders to find solutions that fit their needs and expectations
- Fast POC projects to test reliability of new solutions (real time kernel, fpga, rdma…)
Environment: Redhat linux (6.5/7.2-rt), ptp, TimeKeeper, micro wave, Solarflare, Mellanox, kernel bypass, openonload, perf, strace, ftrace, Bash, Python, Corvil, Overclocking, Ciara servers, HP servers, puppet, ansible,git, jupyter…
Volume: 150 servers.
[06/2012 - 12/2014] LEVEL 3 Support Engineer at INGENICO PAYMENT SERVICES
Inside the Incident and support service of the transactional platform of INGENICO (NOC), being technical leader and senior member in a team of 15 members.
Mostly Troubleshooting incident and investigating problems
- From inputs of all level 1/level 2 support teams of the company worldwide and internal monitoring tools, I was the technical leader in incident resolution and analyze on a platform managing millions of credit card transactions a day, backing the level 2 support.
- I was part of the incident management process and problem management process as technical expert.
- I was part of the Change Advisory Board, writing and supporting change on platform for solving issue or improving the quality of service of the platform.
- 24/24 7/7 full time support with a follow the sun service with Australian team (handover in the morning and late evening for a full day coverage of services).
- Incident could concern all parts of the platform (software issue, system and network issue, provisioning issue, delivery issue) in a very critical environment dealing with payment transaction worldwide.
- In the last year of the mission, i was the main interface with the R&D and developer’s teams, challenging them on various technical topics improving quality of software and deeper bug resolution.
- My vast experience and knowledge of the open source ecosystem leads to very fast and efficient solving of any kinds of troubles, and leveraging positively the stability of the platform and quality of operation. (from on major incident per day, to less than one per semester)
- Participating in recruitment process of other members of the incident and support response team (l1/l2/l3).
Environment ITIL, REDHAT linux, Axis software (Ingenico proprietary payment software), postgresql, haproxy, drdb,clustering, EMC San, CISCO, python, puppet, shellscript, jira, landesk,
Volume: 2000 clients, ~800 servers
[04/2011 - 06/2012] EDENRED (ACCOR France), Platform Manager
Managing prepaid and loyalty program management system of Edenred to End of Life
- System and network administration
- Solving issue
- Managing Tomcat and jboss
- Answering marketing teams requests by advanced extract from Database (Informix, postgresql, mysql)
- Helping customer to migrate to external platform
- Managing the end of life of this platform
Environment: Linux, Informix, Postgresql,Mysql, CFT Axway, 60 servers
Volume: 20 clients, ~60 servers
[01/2009-09/2010] TARANIS (CTO)
Taranis was the historical leader of back office software for TOUR OPERATORS.
Providing application as a service in remote datacenter for +50 international and major tour operators worldwide.
I was team leader of 2 people to maintain and develop the platform.
- Network and system administration
- Full time support to customer, and incident management
- Managing deployment of solution with new customer in Europe
- Managing more than 200 vpn worldwide for customer access to service
- Managing security of the platform
- Datacenter maintenance and managing contracts with different provider (cogent, interoute, orange, free…)
- Migrating customer from standalone solution to hosted application.
- Maintaining +300000 lines bash application framework
Environment: Centos, hpux, cobol, bash, postscript, xml, tomcat,
Volume: 50 clients, ~100servers
[2005-2008] SSII Iniflux Open Source consultant
As a consultant for Iniflux i had several experiences such as :
Project manager at RATP (Paris public transportation) to deploy the new monitor platform. In a team of 6 persons I was in charge of performance audit of the new system which manage the monitoring of the first fully automated metro line in Paris (line 14). I was also interface with developer’s team to report all failure and issue regarding performances. I was the main interface with RATP for any technical issue with the platform.
Environment: Sun Solaris, Redhat linux, bash/ksh, oracle sql server
Network monitoring development for Caisse D’epargne (bank) In a team of 5 people, we developed the monitoring system of the customer backbone based on Nagios.
Environment: Nagios, perl, python,debian, ubuntu
- Research and development for ESCP (business school), I was asked to study the creation of a remote teaching platform based on Video streaming, live chat, and shared desk.
Environment: Jabber, SIP, videolan, xen, LDAP, redhat
- Network Consulting for Transactel (mobile phone operator), auditing their network and improving their firewall policy
Environment: debian, apache, iptables, firehol, cisco-pix, VPN, IPsec
- Security Consulting (Symantec), during 6 months I was the trusted security advisor (TSA) for one of biggest customer of Symantec, during the deployment of a new monitoring solution on its backbone
Environment: Symantec monitoring solution, linux redhat
- Pentester for Telindus, for short period reinforcing their pentester team when the workload was heavy.
- System and network administration for Hosting company Waycom International, Managing their linux servers and customers, Support and incident recovery, Network administration (adsl/vpn).
Environment: Linux, cisco, juniper
[2000 to 2005] Work done while studying
During my study, I managed to work, my working experiences are as following:
- Teaching computer and network security at Orsay University
- Teaching Linux and System Administration at Finance Ministry of Senegal
- Penetration tester at Safetech
- Part time Security officer at Colt Telecom
- Writing a penetration testing methodology for a security company
- Baccalaureat in Science (Mathematical Section in 1997 - Lycée Militaire de Saint Cyr)
- DEUG MIAS Université d’Orsay Paris Sud (Math and CS section in 1999)
- Licence Informatique Orsay Paris Sud in 2003