CTO for Data and AI Storage Solutions at IBM. 25+ years architecting enterprise IT systems across federal, commercial, and public sectors — with a focus on HPC, hybrid, and AI infrastructure, and on unstructured data at scale: file, object, and tape storage across cloud, on-premises, and edge environments.

The repos below are personal open-source work — tooling for large Linux clusters and IBM Storage Scale: benchmarking harnesses, cluster administration, network validation, and install automation.

🔗 cdmaestas.github.io · LinkedIn

• IBM Storage Scale: Working with Abstract Data, IBM Redbooks (SG24-8602), 2026
• IBM Storage Scale System 6000 with NVIDIA DGX SuperPOD Deployment Guide, IBM Redbooks (REDP-5746), 2026
• IBM Storage Scale System Introduction Guide, IBM Redbooks (REDP-5729), 2026
• IBM Storage Ceph Concepts and Architecture Guide, IBM Redbooks (REDP-5721), 2024
• IBM Private, Public, and Hybrid Cloud Storage Solutions (REDP-4873) and IBM Software-Defined Storage Guide (REDP-5121), IBM Redbooks, 2018

• IBM Storage Scale: Working with Abstract Data, Author Session, IBM MediaCenter, 2026
• IBM Fusion Deep Archive and Tape Robotics for HPC and AI, SC25, 2025
• Optimize Your Data Lifecycle, Jeskell digital talk series, 2025
• Roundtable: High Performance Storage (with Patano, Atzkern, Coomer), 2024
• Revamping Data Storage to Harness AI, theCUBE at IBM Storage Summit, 2023
• IBM Storage at Storage Field Day 25, Tech Field Day, 2023
• IBM TechU 2021: IBM Storage Portfolio for Data and AI, 2021
• Spectrum Scale User Group (SSUG) presenter, various sessions, 2018–2025

• CAS for Scale System 6000, IBM Community, 2026
• HPC, AI and Culture: My Trip to SCAHPC Asia 2026, IBM Community, 2026
• Visit to Copernicus LAC Chile, LinkedIn, 2026

Cbench — Scalable cluster benchmarking and testing toolkit. Automates the full HPC benchmarking lifecycle: generating job scripts, submitting them to the batch scheduler, and parsing output into structured results across 31 benchmarks. Originally from Sandia National Laboratories; v2 adds a modern Python toolchain alongside the original Perl. Perl Python

CAP — Cluster Administration Package — Integrates, configures, and manages HPC clusters by bundling common cluster tools (pdsh, genders, Slurm, munge, conman, freeipmi, powerman) behind a unified build and install system with RPM and Debian packaging. Shell

dool — Command-line, real-time monitoring of Linux system resources (CPU, memory, network, disk, load) with a plugin architecture. Forked from scottchiefbaker/dool, a Python 3 successor to dstat. Python fork

pdsh — Multithreaded parallel remote shell client that runs commands across many hosts at once, over ssh, rsh, or Kerberos IV. Forked from chaos/pdsh. C fork

ctrl-data-mgmt — A local file metadata catalog: index the directories you care about, then find what's eating your disk and what you're storing twice. Pure Python standard library, nothing leaves the machine. Python

Scale GUInstall — A single-file web frontend for the Storage Scale Installation Toolkit. Guided, form-driven install, deploy, and upgrade flows with a command preview before anything executes. Ships as signed RPM and DEB packages. HTML Flask

Network Readiness (KOET) — Validates interconnect latency, throughput, and packet loss before a deployment goes near production. Forked from IBM/SpectrumScale_NETWORK_READINESS. C fork

ECE Capacity Estimator — Calculates effective usable capacity for Storage Scale Native RAID (ECE) systems from a hardware configuration, or solves in reverse for the smallest configuration that meets a capacity target. Forked from IBM/SpectrumScale_ECE_CAPACITY_ESTIMATOR. Python fork

Storage Scale MCP Server — An MCP server over the Storage Scale REST APIs and CLIs, letting LLMs query and interact with a live cluster through a structured tool interface. Forked from IBM/ibm-storage-scale-mcp-server. Python fork

Storage Scale Agents — Specialized reference AI agents built on the Storage Scale MCP server, giving LLMs task-focused ways to operate against a live cluster. Forked from IBM/ibm-storage-scale-agents. Python fork

GPFS User Group Tools — Community-maintained tools from the Spectrum Scale (GPFS) User Group for administering and inspecting GPFS / Storage Scale clusters. Forked from gpfsug/gpfsug-tools. C fork

ESS Comprestimator — Samples existing data to project how much capacity compression would actually reclaim before you commit to enabling it. Forked from IBM/ess_comprestimator. TypeScript fork

_{The projects and views here are my own. They are not IBM products or services, and are neither endorsed nor supported by IBM.}