Anthony D'Atri Email & Phone Number

Pittsburgh, Pa, Us

Armonk, New York, Ny, Us

Enabling open-source software-defined storage at scale for enterprise customers

Published photographer, writer, editor. Ceph open-source software-defined storage, SSD, and infrastructure consultant. Expert consultant to investors. Advisor to Further Reach, a non-profit rural Internet connectivity project. Daily contributor to Ceph documentation.

Toronto, Ontario, Ca

Architecting a metrics-enabled software defined storage layer for Kubernetes and beyond, utilizing Ceph RBD and RGW.

Rancho Cordova, California, Us

Enabled and evangelized QLC NAND SSDs as cost-effective alternatives to both TLC and especially HDD storage, with attractive performance. Worked with customers to improve deployments and demonstrate feasibility with TCO modeling. Presented at Ceph Month 2021, upcoming presentation at Cephalocon 2022 about work to enable multiple RGW storageclasses without requiring client action.Intel's NAND BU was spun off as an SK Hynix subsidiary.

West Palm Beach, Florida, Us

Optimized and grew storage services based on Ceph and other technologies, including a global hardware and topology refresh.Successfully launched high-performance Ceph RBD block storage services using NVMe drives and AMD Epyc processors.Designed a modern observability platform based on Grafana and Prometheus, laying the groundwork for replacing Icinga checks with metrics-based alerting.Helped the company improve remote workforce engagement.

Passionate Ceph, storage, and server architect, SRE, and leader.Published author of Learning Ceph, Second Edition.I engage the Ceph software defined storage platform deeply, optimizing architecture and operations. This includes upstream contributions to the Ceph and smartmontools projects as well as architecture and optimization of the full stack from deployment through maintenance, upgrades, expansion, and refreshes.I have passion for clear, complete, and effective documentation. Time and again I've transformed a loose set of incomplete and outdated documents and tribal knowledge into a comprehensive resource enabling operational excellence.I identify multiple pervasive and impactful component firmware defects, engaging vendors for resolution while implementing mitigation and remediation to minimize service impact. This is always with a focus on efficient and safe tooling and procedures that enable 24x7 ops staff to identify and address faults.I scale startup-class services into consistent and expandable deployments, maximizing reliability, efficiency, and serviceability. I balance CapEx and OpEx for wise capacity management, factoring in datacenter resources and server / component lifecycles.I forge partnership between operational and engineering groups, building esprit de corps.I reclaim orphaned resources and drive fraud / abuse mitigation and determent, reclaimed $2M in resources.Team player, enabling others to be awesome by mentoring, project planning, commissioning tools, and supervising operational staff and development efforts.Linux Ceph Software Defined Storage UNIX Workstations Servers Puppet Automation NetApp NAS Solaris Cisco UCS ZFS XFS Icinga Nagios Monitoring Git Jira SolidFire Coraid SAN Perl WWW SSH DNS NVMe TCP/IP RAID Bash BIOS PXE Ethernet NTP Swift Documentation Scripting Development Red Hat RHEL CentOS Ubuntu HP iLO ILOM KVM Ansible IPMI BMC OSX Firmware VMware Virtualization Ubuntu OpenStack Arista Juniper CIMC PagerDuty

San Jose, Ca, Us

• Led the Storage engineering team into the world of DevOps via operational perspectives and practices. I self-embedded in the team and directly inspired the organization-wide SRE program.• Managed and grew 15 petabyte-scale Ceph clusters of up to 750 OSD’s each.• Drove Ceph and SwiftStack server hardware architecture, configuration, and tuning from the application layer down through the operating system to BIOS, BMC, NIC, and HBA for increased stability, performance, and manageability.• Detected and remediated subtle filesystem, disk, and HBA firmware bugs across an install base of thousands.• Guided and engineered change, release, upgrade, and migration efforts without downtime or impact. Forged a constructive relationship with the Change Management team and streamlined high-overhead processes.• Incident response with vendor case co-ordination, root cause analysis, and short / long term fixes.• Fleet management and hardware troubleshooting across continents.• Led complex deployment and upgrade tasks across teams and continents.• Contributed engineering and operational insights both within the team and through the wider organization. Proactively investigated and prevented problems before they could occur.• Leadership through staff development and mentoring, assigning and supervising work done by others.• Introduced the team to on-call duties, processes, and expectations. Devised and managed PagerDuty schedules and escalation policies.• Agile / Scrum / Kanban planning, collaboration, task management, and feature / epic development.• Realized substantial improvements of stability and MTTR through procedure and event guide documents. Offloaded iterative, incident, and maintenance tasks to 24x7 support personnel through mentoring and documentation. This resulted in faster response to incidents with lessened tenant impact.• Thirty-three recognition awards from colleagues and management.

San Jose, Ca, Us

• Administration of multiple UCS-based OpenStack clusters: build/deploy, automation, operation, tenant support, base image creation and management, security, hardware and software troubleshooting and management.• Identified, researched, and drove remediation of multiple potentially disastrous hardware component defects.• Developed custom server firmware upgrade scripting to reduce server downtime from one hour to five minutes.• Passionate documentation of troubleshooting and execution procedures to tame chaos.

London, United Kingdom , Gb

• Led multi-site deployment of Red Hat Enterprise Linux and HP servers with Kickstart. Integrated RHEL / CentOS into the automation environment, migrated Sun / Oracle Solaris systems including hosts previously managed ad-hoc by users. Responsible for hardware budget of $200k and coordination with vendors and Finance for hardware planning, selection, acquisition, and lease management. Managed and monitored storage.• Initiated and managed multiple projects to refresh outdated and difficult to manage network infrastructure at two central and dozens of remote sites. Drove design and deployment.• Managed network resources at 30+ sites, audited and corrected IOS/ Junos /Arista configurations, router ACL’s. • Automated system deployment and configuration of 170 servers via Puppet, migrated from previous management framework. Configured Coraid SAN gear and Symantec NetBackup for backups.