Amazon recently provided* some insights into how Halliburton’s Landmark Graphics unit has deployed its flagship OpenWorks geotechnical database in the cloud. Back in the day, OpenWorks was sold and deployed in-house on an Oracle database. With the advent of Landmark’s DecisionSpace365, a cloud hosted database was required. Enter Amazon Web Services (AWS) Aurora, a ‘MySQL and PostgreSQL-compatible' relational database built for the cloud. This now underpins Landmark’s Data Foundation for the DecisionSpace365 platform. The cloud-native data store can ingest and manage data ‘at-scale’, delivering a ‘fully managed model that can be customized to many industries’.
The move involved a refactoring of the OpenWorks schema and code to both an on-premise PostgreSQL database and into the Aurora host. Taking Oracle out of the equation meant that ‘Landmark could sell its products at a lower price point’. Landmark’s Amanda Smith reported an ‘up to 30%’ performance hike from Aurora, although it’s not clear what Aurora is being compared to.
Landmark is in the process of obtaining service organization controls (SOC) 2 certification for both its infrastructure and solutions—meaning an independent certified public accountant is determining whether the company has the appropriate service organization and controls safeguards and procedures in place to manage customer data. SOC 2 certification involves two separate reports: a Type 1 report, which measures system and control design based on SOC 2 criteria, and a Type 2 report, which measures operational effectiveness. Halliburton Landmark has received a Type 1 report for its iEnergy hybrid cloud solution. Type 2 is scheduled for completion in 2021.
Because Aurora is a fully managed service, AWS is responsible for database operations including manual software patching, setup, configuration, and backup, and completes those tasks faster than the Halliburton Landmark team was able to do when self-managing the original database. Software updates that took weeks or months with traditional databases now only take an hour or two with Aurora.
Aurora has meant that Landmark now operates an E&P solution without needing to be an expert in platform technology. Smith added, ‘We are relying on AWS to provide the tools, and we add our E&P expertise to deliver solutions.’
In a sperate announcement, Canadian geophysical boutique Explor reports ‘breakthrough’ seismic data processing in the cloud using Landmark’s seismic processing** in the Amazon cloud. In the proof of concept trial, multiple benchmarking tests demonstrated ‘an 85% decrease in CDP sort times, an 88% decrease in CDP FK Filtering times and an 82% decrease in pre-stack time migration’. The latter on a 922 gigabyte, 165 million-trace dataset.
* The release includes events that happened ‘eight years ago’ so we apologize on behalf of Amazon for the late arrival of this ‘actualité’.
** More precisely, ‘Seismic Processing, a DecisionSpace 365 cloud application powered by iEnergy on AWS’. Which we take to mean ProMax.
© Oil IT Journal - all rights reserved.