Present at GTC

Participate at GTC with a talk on a GPU-related topic and share your vision with many of the world's brightest minds. Presenters receive a GTC conference pass.


Details Dates
Call for Talks Opens First Week of August
Call for Talks Closes at 5:00 PM (Pacific) Friday, September 27, 2019
Final Notifications for Talks Sent Out Wednesday, December 4, 2019

If you have any questions or would like to reach the GTC Content Team, please contact us.


Submissions must be about your work using GPUs. It can be completed work or work currently in progress. Submissions must provide actual or expected results/accomplishments and must demonstrate a significant innovation or improvement using GPU computing.


The GTC Content Committee will review, rate, and select submissions based on:

  • Technical details and performance results offered and insights to be learned
  • Relevance and timeliness of the presentation to GPU computing
  • Presenter credentials and expertise in the subject matter
  • Presenter public speaking experience(s)


If your submission is focused on a service, technology, or a new product your company is offering, please contact us for information on sponsored session opportunities.


You will be required to provide the following information in your submission:

  • Submitter's Contact Information: Email
  • Session Title: Should be enticing and concisely tell attendees what your session is about.
  • Session Format & Duration: 50-minute Talk (maximum 2 presenters); 50-minute Panel (maximum 5 presenters); 80-minute Tutorial (maximum 3 presenters)
  • Audience Level: 
  • Beginner technical: Session will introduce the audience to the fundamental use cases and challenges of the subject area.
  • Intermediate technical: The audience should have working knowledge of the domain being discussed. Session will typically cover methods and techniques to address specific challenges.
  • Advanced technical: For attendees who have technical background in the subject area. Session will often address specific challenges and complex solutions that may include advanced tricks or techniques.
  • Business/Executive: Broader discussions around AI business models, problem solving, and practical implementation to drive efficiency, improve performance, and enhance the customer experience.
  • Topics: Select a primary and secondary topic if applicable from a dropdown list of pre-approved topic areas.
  • Session Description: Maximum 1000 characters (approximately 200 words) – for use on the GTC website, online scheduler, and mobile app. The session description must explain what the attendee will learn from your presentation and should include any pre-requisite knowledge needed for an attendee to get the most out of your session. It should offer additional details about what will be covered.
  • Extended Abstract & Results: Maximum 3000 characters (approximately 500 words) – the extended abstract will not be published. Provide reviewers with in depth details including examples and actual or expected results/accomplishments from using GPUs to help the Content Review Committee better understand the technical merits of your submission.
  • Presenter's Information: Email and cell phone number (only used for emergency purposes onsite at event if submission is accepted)
  • Presenter Biography: Maximum 100 words
  • Presenter Headshot: No less than 1800x2700px and no more than 2400x3000px. Maximum file size: 100 MB.
  • Presenter Agreement: To be considered for a session the Presenter Agreement must be completed.
  • Permission to Webcast and/or Record: A confirmation of whether or not you agree to be recorded and/or webcast is needed. 
  • Supporting Material: is recommended to assist the Content Review Committee in better understanding your work. Provide supplemental information that supports your submission. Additional materials may include white papers, research work, videos, images, demos, proof of concept, etc.


How to Write a Great Session Title

Your session title is what gets the reader to read the first sentence of the session description. Clearly articulated session titles that have clear learning objectives along with a dash of pizzazz greatly increase the chance that conference attendees will attend the session.

The Title Should:

  • Be USEFUL to the reader
  • Provide the reader with a sense of URGENCY
  • Convey the idea that there is a UNIQUE benefit to attending

How to Write a Great Session Description

The first sentence should describe what the attendee can expect to learn from your presentation (e.g. "Learn about extensions that enable efficient use of PGAS models.") Avoid background your audience already knows (e.g., "Originally designed as graphics accelerators, GPUs have evolved into powerful parallel processors capable of accelerating many compute-intensive applications."). Subsequent sentences should offer more details about what will be covered and why the reader should attend. In general, go for clarity over cleverness.

The Description Should Begin with an Action Word Such As:

  • Learn how to…
  • Explore new techniques in…
  • Hear product experts explain how…
  • See the newest features in…
  • Dive deep into the…
  • Join industry experts for a discussion on…
  • Get the latest information on…

Select the Correct Duration for Your Submission:

GTC is soliciting submissions that provide concrete examples and contain both practical and theoretical information.

  • A 50-minute Talk provides attendees with an expansive discussion of a topic. It covers two or more areas of a broad technical topic or provides a detailed overview or case study of the topic.
  • A 50-minute Panel is an interactive grouping of experts discussing a topic that is best covered from multiple viewpoints.
  • An 80-minute Tutorial provides attendees with in-depth coverage of a technical topic.

Two speakers may be accepted for a 50-minute talk if you can demonstrate the second person is necessary by describing their role in the presentation. 50-minute panels should have no more than five speakers and an 80-minute tutorial no more than three speakers.


Session Title: Faster, Cheaper, Better – Hybridization of Linear Algebra for GPUs

Session Description: Learn how to develop faster, cheaper and better linear algebra software for GPUs through a hybridization methodology that is built on (1) Representing linear algebra algorithms as directed acyclic graphs where nodes correspond to tasks and edges to dependencies among them, and (2) Scheduling the execution of the tasks over hybrid architectures of GPUs and multicore. Examples will be given using MAGMA, a new generation of linear algebra libraries that extends the sequential LAPACK-style algorithms to the highly parallel GPU and multicore heterogeneous architectures.

Session Title: Analysis-Driven Performance Optimization: A New Approach to Determining Performance Thresholds

Session Description: The goal of this session is to demystify performance optimization by transforming it into an analysis-driven process. There are three fundamental limiters to kernel performance: instruction throughput, memory throughput, and latency. In this session we will describe: how to use profiling tools and source code instrumentation to assess the significance of each limiter; what optimizations to apply for each limiter; how to determine when hardware limits are reached. Concepts will be illustrated with some examples and are equally applicable to both CUDA and OpenCL development. It is assumed that registrants are already familiar with the fundamental optimization techniques.


Our runtime designs focus on performance while ensuring truly one-sided progress of communication which is critical for PGAS models. These designs demonstrate significant potential for the users of PGAS models as well as hybrid MPI+PGAS models (as available in MVAPICH2-X) to take advantage of NVIDIA GPUs. The extensions in OpenSHMEM, coupled with an optimized runtime, improve the latency of GPU-GPU shmem-getmem operation by up to 90%, 45% and 42%, for intra-IOH (I/O Hub), inter-IOH and inter-node configurations. The proposed extensions and the associated runtime reduces the latency of 4 bytes Put to 2.7us from GPU-to-GPU. The proposed enhancements improve the performance of Stencil2D kernel by 65% on a cluster of 192 GPUs and the performance of BFS kernel by 12% on a cluster of 96 GPUs. As part of the talk, we will use benchmarks from the popular OSU micro-benchmark suite and application kernels to demonstrate how to use the new extensions and extract performance benefits from the associated runtime designs.