Oscar Luiz Silva Neto
Verified Expert in Engineering
Database Administrator and Developer
Oscar has been a database administrator for over 25 years—acting as a consultant and customer—meeting each project's demands and expectations to maximize the financial benefit of the customer's investment. Oscar specializes in data and how it will be computed, stored, and made available to client applications to continuously and effectively provide the necessary information. His experience is supported by 15 years as a consultant and freelance implementer for HPE, IBM, and Oracle.
Portfolio
Experience
Availability
Preferred Environment
Architecture, Virtualization Technology, Disaster Recovery Plans (DRP), Database Architecture, Data Migration, Azure, SAP, Oracle DBA, Grid Computing, SQL Server DBA
The most amazing...
...project I've worked on was a SAP migration from HP-UX to Linux of 70TiB with two hours of downtime.
Work Experience
Trainer
Savant IT Services
- Developed specific onsite or remote training solutions for clients.
- Created classes using Zoom and labs at Azure with 100% clients' satisfaction.
- Focused on organizations whose students are geographically dispersed.
- Gave online lessons about Oracle RDBMS and MS SQL Server regarding day-by-day DBA activities, monitoring, performance, and troubleshooting.
Principal Database Administrator
KnowHow IT Services
- Served as a database solutions architect, creating or migrating more than 100 installations, for MS SQL Server, Oracle (RDBMS and MySQL), and EnterpriseDB (PostgreSQL).
- Ported more than 50 applications from legacy platforms (AIX, HP-UX, Solaris, OS/390) to new platforms (MS Windows and Linux) or the cloud (Azure).
- Created five high-performance clusters, for AI, machine learning, computational fluid dynamics, and multidisciplinary structural analysis.
- Worked on projects concerning database exabyte storage backup and disaster recovery, performance, and administration, focused on immediate service recovery.
- Worked on projects involving data security focused on the Brazilian GDPR: General Data Protection Regulation (Lei Geral de Proteção de Dados).
- Planned more than ten disaster recovery plans and their testing.
- Created database administration plans for local support teams that focused on daily activities and targeted Oracle RDBMS/MySQL and MS SQL Server.
Infrastructure Consultant
HPE Client - Protected by NDA
- Migrated 100 TB of virtual machines between VMWare clusters without application downtime.
- Changed the configuration of virtual machines to address the best possible performance, with I/O times reduced by up to 90%, using VMWare Powercli scripts, requiring downtime only for VM reset.
- Tiered disk storage configuration, where the hot data was on NVM/SSD drives and the cold data was on NL drives, in order to maximize the investment made by the hospital.
- Designed and implemented a complete backup infrastructure using Veeam and HPE StoreOnce, with integrated backup to storage unit snapshots and data transfer directly via the SAN without burdening the network or the hypervisors.
- Designed and implemented a complete disaster recovery solution using VMWare SRM + Veeam VBR, where business-critical VMs were replicated via SRM and non-critical ones with VBR. We achieved an RTO of only 30 minutes with an RPO of 15 minutes.
Disaster Recovery Plan Specialist
Client
- Developed a disaster recovery plan, analyzing the points that go from infrastructure to crisis management—detecting the incident, triggering the crisis committee, and communicating with the internal public and its customers.
- Defined the RPO and RTO carried out with the business and planned the feasibility with the technical team; also created a mitigation plan to cover the infrastructure gap.
- Created a rehearsal plan, foreseeing the creation of a "bubble" using the secondary infrastructure, in a way that does not affect the production.
- Developed an activation plan, detailing the activation commands in a row, so that anyone with some technical knowledge can execute it.
- Advised the technical team to create the secondary infrastructure (in progress).
Database Administrator
Client
- Performed a complete database solution with master-node and read-only replicas, achieving 1.4 million IOPS.
- Handled the full backup implementation with pg_backup, DRDB, snapshots, and deduplication—providing an RTO and RPO closer to zero.
- Created a complete monitoring solution with Zabbix as the framework and Grafana as the presentation layer—giving customized views of each solution component (performance, network, queries, replication) to the ongoing team.
Database Administrator
Toptal K-12 Education Client
- Explored and tuned SQL queries to save resources from PaaS solutions which helped the clients to spend less money.
- Performed a complete database health check before moving to the cloud—pointing out what needs to be fixed in an ABC priority list and avoiding the troubles after cutover.
- Coordinated with the development team to review the application logic and achieve the best approach to data.
Developer (Unix and Storage Support)
Banrisul
- Administered and supported more than 20 high-end Unix systems.
- Created an active-active DB Metro Cluster using Oracle RAC Extended Clusters, with HP-UX, a stretched SAN, and redundant storage array systems.
- Implemented a large SAN with Cisco MDS switches, including Metro SAN between sites.
- Implemented large databases backups using IBM 3494 silos.
Database Administrator
InterOp
- Served as an HPE (former HP) C&I solution implementer for telecom and utility industries, with more than 20 projects for call details record (CDR) data manipulation and treatment.
- Designed and implemented a database infrastructure for electric power distribution border measurement systems in a large Brazilian utility company.
- Designed and implemented a database infrastructure for customer care and billing systems for prepaid telephony, especially Amdocs CCBS and Ericsson CDR Treatment (former Ericsson Hewlett-Packard Telecommunications).
- Implemented four SS7 networks at Telefonica Brasil with HP France.
- Designed and implemented an OLAP DB from network routers and switches to analyze performance, availability, and audit billing.
Database Administrator
Calçados Beira Rio
- Performed a complete database redesign with an online backup and resource isolation in Oracle 7.
- Implemented ten databases that were distributed and replicated using private data communication lines not to depend on a central node.
- Implemented parallel computing with Oracle Parallel Server.
Database Administrator
Telet (now Claro Brazil)
- Worked at the startup that became the largest mobile phone network in Brazil, creating the entire database infrastructure from scratch.
- Focused on customer care and billing systems with 100,000 new customers every month.
- Constructed disaster recovery plans for all systems required for business operations.
- Planned and deployed a large OLAP database for the anti-fraud system.
- Planned, deployed, and maintained a CDR DB with ETL tools from Oracle.
Database Administrator
Dana
- Implemented a highly available database system for the automotive industry in order to meet JIT (just in time) and Kanban specifications.
- Administered an Oracle manufacturing database with a corresponding application.
- Migrated data and integrated it between heterogeneous systems (AS/400, Oracle, AIX, and OS/390).
Experience
Database and App Disaster Recovery Plan
Used Components:
• Large databases instances on Sun Fire E25K Servers
• Oracle RAC, E-Business Suite (Financials) database
• TIBCO ESB
• SAP R/3
• Geo-dispersed cluster (more than 1,000 km)
• 4-hour RTO/8-hour RPO
• Oracle Dataguard Maximum Performance mode
HCI Infrastructure for Databases
Project Details:
• Incorporated 40 HCI HPE synergy large-size nodes.
• Implemented VMWare ESXi 6.7 Enterprise Plus.
• Built a 200TB with zero downtime backup and a zero recovery time SimpliVity OmniCube backup.
• Integrated with Veeam Enterprise Backup for geo-redundancy and to avoid ransomware contamination.
• It had a 2-week total project timeline.
SAP R/3 ECC Platform Migration
Project Details:
• Implemented it with minimum downtime, only two hours.
• From HP-UX to Linux, I used Oracle Cross-Platform Transportable Tablespaces (XTTS) and RMAN.
• Migrated it from a single-instance failover cluster (MC/ServiceGuard) to an active RAC cluster.
• Took it from cooked data files to an ASM normal redundancy with dual-storage arrays.
• Started six weeks before the cutoff date, with a QAS environment for system approval and management, backup, and performance monitoring routines.
Results:
The process saves more than USD $500,000 in license and labor costs and an unpredicted value in the maintenance of legacy hardware for the next five years.
SAP Hana ECC HA Solution
Project Details:
• Transformed it from an 8TB database to a 12TB one now.
• Enabled local and geographic high availability using a 3-node TDI ECC solution.
• Implemented geographic high availability using a 2-node TDI BW scale-out solution.
• Developed an online, array snapshot-based backup solution that was fully integrated with Veritas Netbackup and HPE StoreOnce.
Results:
We reached our recovery time objective of five minutes with 10,000 simultaneous users.
Telecom Appliances Centralized Backup
There is no tool on the market to centrally backup telecommunications transmission and switching equipment. Each manufacturer (Ericsson, Huawei, Alcatel, Cisco, H3C, HPE, Oracle) has its backup method, which as a rule, is simple and does not meet the minimum demands for data protection, such as retention, confidentiality, tracking, and, mainly, ease of recovery.
The Solution:
I assembled a tool to back up this heterogeneous range of telecom devices.
Used Components:
• Bash and KSH scripts to command backups via native tools
• Microfocus (former HPE) Data Protector, to orchestrate backups and restores
• HPE StoreOnce, as a storage destination, providing deduplication, encryption, and geo-redundancy
• A PostgreSQL database to inventory the devices and save the backup method
Results:
More than 20,000 heterogeneous devices started to have a centralized, traceable, and secure backup at the end of the project, meeting regulatory needs.
High-performance Cluster for Computational Fluid Dynamics and Multidisciplinary Structural Analysis
Project Details:
• Paired each cluster with 128 HPE BladeSystem nodes and 2048 high-clock Intel cores.
• Implemented Infiniband non-blocking fat-tree topology at 56 Gbps.
• Installed NetApp Fillers as a scratch area.
• These were the first Brazilian clusters that were used Open MPI as a message-passing interface.
Results:
It was 2.5 times faster than the previous solution (a remote HPC located on the US) with a quarter of the cost in nodes and 2048 high-clock Intel cores.
HPC for Fluid Simulation Software
Used Components:
• Four HPC compute nodes with 228 processing cores
• Two VMware Horizon VDI Servers, each one with two shared GPUs, controlled by NVidia Grid
• Eight high-performance virtual desktops performing more than 60FPS
Results:
The customer saves more than USD $500,000 in hardware and software infrastructure. It worked perfectly while working during a COVID-19 remote office schedule.
Data Center Migration
Project Details:
• IP addresses that are different from the source to destination data centers; this was a task that involved more than 200 applications to be reviewed, solved with NAT and IP forwarding techniques.
• More than 200TB (at 2009) were replicated, using different kinds of approaches, like array and database replication, backup/restore techniques, and a complete reinstall of application servers.
The Results:
The project was delivered on time, without any loss of data and the customer saved more than USD $1 million per year in real estate costs.
Cloud DRP for On-premise Databases
Project Details:
• 100% virtualized environment with VMWare ESXi
• PROD, DEV, and QAS environments were separated into different VMs
• Databases were grouped in multitenant environments of the same category
• Used Azure ExpressRoute to provide network transparency
• QAS and DEV moved to Azure to keep the site warm
• PROD was replicated using SRM to automate the activation of the site
• All data was encrypted at the destination
The Results:
• Fully integrated with satellite systems, including PACS
• New backup strategy focused on the cloud
• Created activation and rehearsal scripts
• Only two rehearsals performed to validate the procedures
• RPO: 3 hours
• RTO: 30 minutes
Database Infrastructure for OTT Media Service
Project Details:
• Largest telecommunications provider in Brazil wants to deliver multimedia services via streaming over the internet
• Initial expectation of 20 million users
• Workflow and device management, dashboarding, reporting, billing, and media content management via databases distributed in VMWare Farms, fully redundant with geo-replication in several regions of the country
• MongoDB and MariaDB as databases
• Array-based snapshots because there would not be enough time to restore from tape
Results:
• From the drawing board to full production in 25 days—five days before the deadline
• 78 production VMs and another set for QAS
• Zero downtime backups
• An undisclosed amount due to the end of the loss of revenue with illegal TV broadcasting
Geo-redundant Oracle SAP Cluster
Project Details:
• Six databases on twelve Oracle RAC nodes on Linux
• 25TB of data
• Data replication via IP network
• RPO close to zero
• one hour RTO
• Zero RCO, that is, the databases must be consistent with each other
Solution:
• Asynchronous Oracle Data Guard replication
• Replication management via Oracle Enterprise Manager
• Data Guard Broker to maintain consistency across databases
• Complete set of documentation for the switchover in case of rehearsal and activation
Database Datacenter-centric Migration
In this migration, the challenge was changing the IP address of database servers because many applications did not use name resolution (DNS), only via files from hosts or even by the IP address itself.
For that, it was necessary:
We documented all the applications and the way they connected and change from IP or host a file to DNS. We also provided a mechanism so that if any application could not be changed it could continue to access the database and log it somewhere so that it could be tracked.
The solutions were:
We used a CMDB to document the database, that captures all information needed and a NAT at the edge of the network to redirect calls and track usage.
SAP BW/4HANA Scale-out Cluster
Project Details:
• Solution: HPE Superdome Flex scale-out cluster
• Storage: Dell/EMC VMax
• Database size: started with 25TiB, today it has more than 100TB
• High availability with HPE Serviceguard for Linux
Data Warehouse (DW) for Credit Histories Using Microsoft APS
Project Details:
• Solution: Microsoft PDW
• Storage: HPE 3Par
• Database size: started with 5TB, today it has 30TB
• Implementation of Infiniband as the interconnection between nodes
Code Tuning to Save PaaS Resources and Money
It was verified that there was room to improve the application queries, resulting in response times of up to 1/200 less.
We transferred knowledge to the customer of how the data is accessed and what is the best way to do this so that he could make the right adjustments in his applications; in addition, it was explained how PaaS solutions scale in terms of I/O so the best solution is always to "consume less."
Skills
Languages
Bash Script, SQL, Bash, Stored Procedure
Libraries/APIs
Microsoft HPC
Tools
VMware, NVIDIA Grid SDK, Hyper-V, Oracle Exadata, Cluster, Oracle RAC, Cron, Pacemaker, Zabbix, Zoom, Control-M, Distributed Replicated Block Device (DRBD), Grafana, Amazon CloudFront CDN, PowerCLI, Data Guard, OpenFOAM
Paradigms
High-performance Computing, OLAP, Database Design, ETL
Platforms
HP-UX, Linux, Unix, Oracle, Windows, KornShell, Oracle Enterprise Manager (EM), Azure, Oracle Database, Ubuntu, Red Hat Linux, AIX, Solaris, HP OneView, SAP HANA, Citrix, Azure PaaS, Amazon Web Services (AWS), AWS Cloud Computing Services
Storage
SQL Server DBA, Oracle DBA, HP Data Protector, HP StoreOnce, HP 3PAR, EMC VMAX, PL/SQL, Oracle PL/SQL, Databases, PostgreSQL, Microsoft SQL Server, Database Architecture, Oracle Database Tuning, Database Replication, HP BladeSystem, HP Servers, SQL Loader, Database Backups, Database Administration (DBA), Database Modeling, Relational Databases, Database Migration, Oracle Multitenant, Configuration Management Databases (CMDB), SQL Server 2014, MySQL, MariaDB, MongoDB, Microsoft Parallel Data Warehouse (PDW), SQL Performance, Azure SQL, Database Performance, Distributed Databases, Oracle 11g, IIS SQL Server
Industry Expertise
Telecommunications, Banking & Finance, Healthcare, Automotive, Social Media, Insurance, Retail & Wholesale
Other
Oracle RMAN, SAN Brocade, Cisco MDS, Grid Computing, Clustering, InfiniBand, Unix Clustering, Unix Shell Scripting, Storage, SANs, Backup & Recovery, Disaster Recovery Plans (DRP), Business Continuity & Disaster Recovery (BCDR), Disaster Recovery Consulting, Database Optimization, Database Table Optimization, Oracle Data Guard, Server Migration, Data Migration, Telecom Business Support Systems (BSS), Telecom Equipment & Solutions, Industrial & Manufacturing, Hyperconvergence, SAP ECC, Scripting, Auditing, HPCC Systems, Website Migration, SRM, Manufacturing, Consulting, Virtualization Technology, Data Warehouse Design, Data Warehousing, SQL Server Administration, Architecture, Performance Tuning, Monitoring, Data Protection, Indexing, Enterprise, Shell Scripting, Writing & Editing, High Availability Disaster Recovery (HADR), SSH, SFTP, System Administration, Windows Subsystem for Linux (WSL), IT Support, Email, SAP, Veeam, NetBackup, IBM Storage, Microsoft Azure, Oracle EBS, Oracle Apps, Logistics, Education, Technical Training, Military, Heartbeat, Utilities, Transportation & Shipping, Networking, Fluid Dynamics, Structural Analysis, Aviation, Design for Manufacture & Assembly (DFMA), iSCSI, IP Networks, Aerospace & Defense, ETL Tools, Data Analysis, IBM ODM, Cloud Architecture, Virtual Desktop Infrastructure (VDI), Aruba, Machine Learning, Cloud, Mining, Nokia Telecom Application Server (TAS), Cisco, Government, Construction, TIBCO, SAP BW on HANA, Alcatel, NAT, Picture Archiving & Communication Systems (PACS), Access Control, IPTV, Over-the-top Content (OTT), Startups, K-12 Education, IaaS, Supercomputers, Fault Tolerance, Servers, Credit Collection, Credit Industry, YugabyteDB, Sharding, Amazon Route 53, eCommerce, Images, MCDBA, Battery Management Systems, Electric Vehicles, TIBCO BusinessWorks, Cloud FinOps
Certifications
Battery Energy Storage System
Weg
AWS Partner: Cloud Economics Accreditation
AWS
YSQL Development Fundamentals — Associate
Yugabyte
HPE Tech Pro Mission Critical Servers
Hewlett Packard Enterprise
HPE Tech Pro HPC Supercomputing
Hewlett Packard Enterprise
HPE Tech Pro Data Protection
Hewlett Packard Enterprise
Hybrid-cloud Solution Archited
Hewlett-Packard Enterprise
Solutions Certification
Nvidia
Database Administrator
Microsoft
Certified Professional
Microsoft
Database Administrator
Oracle
Storage Technical Support
Hewlett Packard Enterprise
HP-UX Certified Professional
Hewlett-Packard Enterprise
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring