Moore’s law (an exponential growth in the number of transistors per integrated circuit) simultaneously delivers both reduced cost and improved capabilities, but most managers focus more aggressively on cost reduction than on capability improvement—and understandably so, as costs continue to drop dramatically for products used in established work processes. This approach leverages Moore’s law to do today—at a lower cost—what your business could do yesterday, and has been doing for 20 years.

But inevitably, technology innovators develop and deliver breakthroughs that can fundamentally change the economics of companies dependent upon information technology (IT) to drive their business and profitability. So at some point every industry needs to step back from the focus on reducing cost to ask the question: Can I do things differently in my business today that will substantially improve performance, and materially improve it much more aggressively, than I could ever achieve by saving another 10% on a cheaper personal computer? By grasping new technologies, often at the same or reduced cost compared to legacy systems, the industry can reap both financial savings and vastly increased computational capabilities. Energy companies that have decided to take advantage of both curves of Moore’s law are enjoying increased efficiency, production, and profits.

Improving Performance
In the mid-1980s, the geophysical work process changed from working with paper sections and maps to interpreting seismic data and automatically creating the resultant maps on computers. Although the movement to computers was a significant breakthrough because it allowed data to be viewed and understood much more quickly, the memory capacity of those early computers was so limited that the industry went from using multiple seismic attributes in the interpretation to only one: amplitude. The task of loading a second attribute to the process was difficult and time-consuming, and there was no effective way of simultaneously viewing multiple attributes. So additional attributes often were not incorporated. Fast forward 20 years, and there is an established standard work process in which seismic interpreters, having entered the workplace after the transition to computers, work predominantly with only a single attribute.

The computing capabilities of 2005, in contrast to 20 years ago, can take us back to where we were before the ubiquitous introduction of work stations, where multiple simultaneous attributes can be used in the standard work process but are now visualized and analyzed together in a common volumetric digital image. For example, last year, SGI and Landmark Graphics conducted a technology breakthrough using a data set loaned by Marathon Oil. Using Landmark’s Geoprobe interpretation software running on an SGI visualization system, the team successfully demonstrated how a 400 gigabyte volume could be visualized and analyzed in real time. This was a multiattribute volume that allowed the interpreter to look at the amplitude volume corendered with a continuity volume, providing a high-resolution view of key faults and stratigraphic discontinuities in the volume.

Next-generation products bring this multiattribute corendering ability into the computing environment affordably and enable potential breakthroughs in work processes because of a transition to the Linux operating environment and the replacement of proprietary components by industry standard processor and graphics card technologies (Fig. 1).

Fig. 1—Innovations can deliver breakthroughs that improve the economics of companies dependent upon IT to drive their business and profitability.

Mimicking the work process typical of paper-section analysis, multiple attributes can re-enter the interpretation, and other categories of data such as facilities, geology, wells, and real-time reservoir performance measurements can be incorporated as well. Because of the scalability of these new systems, they can be loaded with all of the architecture and mechanicals of the fields’ wells. All of these data, either historical or predictive, can now fit into the model in the active shared memory of these machines, scaling up to 4 terabytes of memory. This expandable memory delivers a whole new set of capabilities, enabling better science much more quickly and better constrained, resulting in improved exploration success factors and improved reservoir recovery and performance.

The installation of large-format visualization theaters evolved from novelty, high-tech presentation rooms to ubiquitous corporate collaboration environments. There are nearly 200 visualization theaters installed around the world in energy companies. In recent months, companies have been moving and adapting these visualization systems from the discovery to the recovery process. This is an important change because companies are adapting things that they know how to use well and bringing much more affordable versions into the office to bring their engineers and operations personnel together into a common environment. Instead of being out on the platforms in the middle of the North Sea, operations people can now sit in an office adjacent to the field engineers and scientists who have access to all the new data, allowing them to collaboratively make operational decisions for the fields.

Analysis in a Single View
The new scalable graphics computers provide the ability to consume ultralarge quantities of seismic or reservoir data in a single view. In the past, mostly because of small-capacity workstations, it was impossible to analyze large geographic areas because the data would not fit in the computer. The large area was broken up into pieces small enough for the computer and handed out to multiple users, who did their work independently of each other and then tried to bring it together at the end. It was the only model available for ultralarge projects, but the parts may not fit together when each piece of the analysis is complete. To repair the mismatches at the edges required either a lot of reworking or, in some cases, reinterpreting some or even most of the elements to make a consistent and internally coherent model.

Today, that information can come together in a single interpretation session by using advanced computers and modern software to enable an individual or team to have access to 100% of the data all at the same time—including the different attributes, reservoir models, and geological interpretations. By approaching it as a single entity, there are no discontinuities at the edges of the work, and large-scale regional geological trends and anomalies can be investigated, the first important phase of exploration.

Streamlining how people work and making a huge jump forward in the amount and kinds of data they are able to look at has the effect of potentially improving exploration success rates and improving recovery from the reservoir as well as reducing costs of operations. Consider this: There are approximately a trillion barrels of proven oil reserves in the world today, with an average recovery rate of 30%. If the industry doubles the recovery factor on known oil fields around the world, it would add an additional trillion barrels of producible oil. At today’s price of U.S. $50/bbl, that is approximately $50 trillion of added global reserves. Just as important, this provides the ability to sustain production profiles longer into the future and shift Hubbert’s Peak further into the future yet again.

That point in the future at which we start to see significant declines in production has seen numerous forward revisions, all because of major breakthroughs in technology such as 3D seismic, horizontal drilling, deepwater production, and subsea completions. If the industry uses these new visualization technologies to achieve a four- or five-fold increase in exploration success and is able to dramatically improve recovery from reservoirs, the world economy will continue to enjoy access to a vital supply of cheap energy until it transitions to new types of energy.

Increasing Exploration Success
On the exploration side, ExxonMobil is on record that it has improved recovery rates from 10% to nearly 50%. There are 1,450 active drilling rigs in the world today. These wells cost U.S. $10–15 million apiece, and maybe the discovery is 30 million bbl. If the industry can increase the recovery rate by four times, that is an additional U.S. $10.5 trillion in reserves and $70 billion in dry-hole savings per year. You can see that the numbers get very big very fast as the production and discovery rates climb.

Energy customers around the world are using lower-cost, higher-capability visualization systems with demonstrated substantial economic benefits. A Latin American customer recently streamlined its work process from months to weeks, brought its asset teams together into a common environment, and was able to drill 14 fewer wells in a particular field. Because it was spending U.S. $15 million per well, that resulted in savings of more than $200 million from one operation. Similarly, a customer in Norway reduced drilling costs by $20 million to $40 million and, by sharpshooting the position of these wells, was able to add an additional $375 million worth of oil from these wells. In conclusion, the good news is that IT managers can get a lot more capability, and also spend a lot less money, without having to sacrifice cost for improved capabilities.