The BREEZE Remote Modeling System is designed to deliver significant performance improvements on large AERMOD model runs.
Improvements will be observed for scenarios that containing any source type or model option. Estimating how long an AERMOD model run is not an exact science. There are a number of variables that will affect the performance, including, but not limited to:
- the Control Options selected - certain Control Options will enable additional routines, thus additional computations.
- the meteorological data - different meteorological conditions require different algorithms (e.g., convective boundary layer vs. stable boundary layer).
- terrain processing - computing dispersion in complex terrain requires addition processing. AERMOD will computer a critical dividing streamline height for each hour for each receptor. The combination of the receptor and meteorological conditions for a particular hour will determine which algorithms will be used.
- building downwash - when building downwash is considered, additional computational routines are triggered.
- use of external files - a lot of reading and writing of files will reduce performance. The BRMS does not permit the creating of ASCII POST files due to their potential large size. Binary files can be created.
This page describes how you can achieve phenomenal runtimes with the BREEZE Remote Modeling System and how fees are assessed:
How it Works
The BREEZE Remote Modeling System (BRMS) for AERMOD operates on a massively parallel computer cluster that combines the processing power of multiple multi-core computers. Multi-core computers have more than one processing unit, or CPU, on a computer chip. Each core can be viewed as a separate computer capable of performing independent calculations. Expert air quality modelers are anxious to save time and money by utilizing scalable, high- performance computing (HPC) applications to process the numerically intensive algorithms and data often associated with dispersion models. The Remote Modeling System utilizes the BREEZE AERMOD Parallel Fortran application, which debuted in June 2007 at the Air and Waste Management Association’s (AWMA) annual conference.
The Need for Speed
As air quality modeling software and associated preprocessing applications become more complex to address the physical and chemical processes occurring in the atmosphere, multi-core computers or clusters are becoming more attractive. Software designed to utilize the full computing power of these systems allows modelers to complete model runs in a reasonable amount of time. Currently, serial versions of the AERMOD air dispersion model available from the U.S. EPA can take anywhere from a few seconds to complete for a very simple scenario, to over a month or longer for a complex multi-source, multi-receptor scenario (e.g., SIP or PSD-related scenarios). Although most modelers now have computers with at minimum dual-core CPU, the dispersion modeling applications available from EPA do not utilize these capabilities. This effectively reduces the return on investment for a multi-core CPU. To take advantage of multi-core computers, the user requires applications that are specifically designed to run in that environment. True parallel computer applications do this by sending each core/CPU its own set of instructions simultaneously.
The U.S. EPA serial-version AERMOD dispersion model may be required to make thousands of computations for each hour of meteorological data processed. Each computation is completed sequentially, one at a time. Serial processing can be thought of as if each computation were to line up and file through a single door (where the door is the CPU). Parallel programs, such as BREEZE AERMOD Parallel, essentially open more doors (i.e., CPU’s) by sending multiple computations simultaneously to multiple cores/CPU’s.
The U.S. Department of Defense (DoD) recognized Trinity’s ability to express numerical algorithms at a higher level of abstraction than those currently employed by publicly available models and contracted Trinity’s BREEZE software developers to develop high performance computing parallel versions of applications such as CALPUFF, CALMET, and SCIPUFF to run on standard PC’s utilizing the Windows or Linux operating systems. To address DoD’s need for more efficient modeling systems and ensure that these modeling systems complied with EPA’s Guideline on Air Quality Models (e.g., no changes to model algorithms), Trinity initiated Project HPCM, the High Performance Computing Modeling project. Project HPCM’s goal was to identify parallelization techniques that not only met the needs of the DoD but that could also be used more widely to produce results identical to those produced by EPA’s public models. A high priority objective for developing parallel versions of these dispersion models was to generate unobtrusive code which, when applied to an existing serial program such as AERMOD, generates a functionally equivalent parallel version (i.e., identical results).
The Challenge
Until recently, parallel programming has been largely restricted to either purely academic activities or to super-computer systems. Advances in personal computer technology have made the availability of multi-core/CPU computers mainstream; however, many authors of serial processing codes have been slow to adapt their applications to the capabilities of these systems due to the cost and level of effort involved to properly parallelize code. A significant challenge associated with developing efficient parallel computer models is the routing of individual computations to multiple processors without affecting model results. Achieving identical results to EPA’s public models was the top priority for developers of BREEZE AERMOD Parallel.
BREEZE AERMOD Parallel developers were charged with not only developing a version of the AERMOD application that optimized the use of multi-cores/CPU’s (e.g., Intel’s CoreTM Duo processor), but also to extend the processing across multiple, networked computers, such as in a computing cluster. A cluster refers to a network of independent computers linked together to form a unified system with the objective of completing a common task. Clusters are often built using commodity hardware on a private network with open source operating systems (e.g., Linux), thereby making available powerful computational resources to those with even the most modest budgets. Simply building a cluster does not afford a modeler the advantages of a multi-core/CPU system. The design of the software used on the system is critical. Programs that are poorly architected or communication-intensive can run even more slowly on clusters or multi-core/CPU machines than their serial counterparts.
A significant number of environmental regulatory agencies worldwide currently run numerically intensive applications such as AERMOD and CALPUFF on serial computers and are faced with roadblocks that prevent them from taking advantage of parallel architectures. One issue is that users who could benefit most from parallel processing may lack the knowledge necessary to extract optimal performance from their available computing resources. Moreover, they tend to have large legacy systems running on serial machines, and re-writing this legacy code can represent an enormous cost.
The generation of good parallel code is not a simple task. Developing optimized code requires substantial skill and a significant mind-shift from how serial code is prepared. BREEZE developers have produced parallel code that can provide substantive value to dispersion modelers. The following equivalency demonstration is provided to communicate the equivalency between BREEZE AERMOD Parallel and the EPA’s public domain AERMOD application.
There are a number of payment options you can subscribe to:
- Pay as you go - call or email us to purchase a one time use BRMS key. You will be able to submit one model run to the BRMS. The run can be submitted from BREEZE AERMOD GIS Pro V6.2 or greater, or via the Web submission form.
- Monthly invoicing - call or email us to sign up for montly invoicing. You will receive a key that allow you and other modelers you provide the key to, to make as many runs as you like. At the end of each month, you will be provided an invoice detailing your charges.
- Reserve BRMS time - There are times when you absolutely have to have model runs completed by a certain time. Because there is the possibility that others submit there model runs ahead of yours, you may have to wait a period time before your model run completes. You may reserve a block of BRMS time in advance and be assured that the system is ready and available when you are. This option requires a non-refundable payment of the cost to reserve half the scheduled period. Once your model run or runs complete, you will be invoiced for the balance, if any.
Wait less. Get more done.
When quantifying the impact of improved model runtime efficiency, there are generally two types of savings – hard costs and soft costs. Hard-cost savings are tied directly to the amount of money saved from the impact. Soft-cost savings are typically tied to improved staff productivity.
With the BREEZE Remote Modeling System, you'll be able to complete what would normally take days or even weeks in only a few short minutes or hours. Imagine how much more productive you can be. Consider the benefits:
- Get results back in a fraction of the time - We're not talking 2, 3, 10, or even 20 times faster. We're talking about 100 times faster when you have multiple, large, AERMOD model runs. Getting results back faster allows you to respond to your client faster so that you can move on to other projects.
- Be more competitive - when submitting proposals, you can increase your chance of winning project work by being able to provide results to the client much quicker. To them, time is money also.
- Make model runs anywhere - The BRMS is perfect for the busy executive or consultants on the go. Start your model run on the BRMS, catch your flight or hop in the car, and get your results when you arrive at your destination (perhaps your clients site). There's no need to worry about your computer's battery, and the model run dying at 30,000 feet.
- What-if analyses - There's a lot to be said for doing a job right the first time. Because you can complete model runs so much faster, you can perform a number of additional model runs that evaluate different source characterizations, receptor configurations, or meteorological data. Presenting your client model results that cover all the possibilities show thoroughness that in the long run can save them money.
- Reduce errors - Some modelers spend an incredible amount of time trying to break up a model scenario and run it across multiple computers. Those error prone days are gone! The amount of labor that is required to create the files, generate the additional output, track the output, and piece them all together in the end isn't worth the effort. Plus, there are too many places where data corruption and errors can be introduced that would require you to have to do it all over again.
- No proprietary software needed - Modelers can upload their model scenarios from any computer with an Internet connection. All you need is a web browser. What's more is that the BRMS accepts any U.S. EPA AERMOD formatted input file. You do not have to have the BREEZE AERMOD graphical user interface (GUI) to create your input scenarios. Simply zip up your project and upload it to our system.
- Reduced dependency on hardware - The BRMS eliminates, to a large degree, your need to purchase, maintain, and upgrade computing hardware and software. Because the model run is submitted to the BREEZE modeling system and run on our computers, you don't need to worry about those extra costs. Don't forget the additional labor costs associated with their maintenance and upkeep.
AERMOD Version 07026 Equivalency
The BREEZE AERMOD Remote Modeling System executes the U.S. EPA AERMOD Fortran executable that has been parallelized for performance across multiple processors (AERMOD Parallel) by BREEZE developers. No changes to Fortran algorithms have been made. In fact, results generated by the BRMS and BREEZE AERMOD Parallel are identical to those generated by the U.S. EPA AERMOD Fortran executable.
As part of the product development process, Trinity staff performed hundreds of AERMOD model runs using both the AERMOD Fortran executable available from the U.S. EPA SCRAM Web site (http://www.epa.gov/scram001/dispersion_prefrec.htm) and the BREEZE AERMOD Parallel Fortran executable developed by Trinity Consultants, Inc. The U.S. EPA version number for both of these executables was 07026.
Trinity has made available to all U.S. state and federal regulatory modelers the BREEZE AERMOD Parallel modeling application to assist in their equivalency demonstration. In addition, Trinity has made available, as requested by OAQPS, model runs completed with the SCRAM version of AERMOD and BREEZE AERMOD Parallel. Model scenarios include the test cases available for download from the AERMOD model page at the EPA’s SCRAM website (http://www.epa.gov/scram001/dispersion_prefrec.htm) and those described in a related memo ( http://www.epa.gov/scram001/7thconf/aermod/evalreadme.txt).
An equivalency document is also provided listing the differences between the model output list file generated by the SCRAM version as compared to the BREEZE AERMOD Parallel version. This document demonstrates that the only differences in the two results sets were the time stamps at the top of each output page.
The data files for the test runs can be downloaded using the links provided below. The results of the file comparison are available here.
The zipped data sets identified below contain the following files:
.SFC and .PFL model ready meteorological data files. The .SFC file was modified by changing the version number from 02222 to 06341 to make it compatible with AERMOD (i.e., allow AERMOD to process it without error).
AERMOD input files for both the EPA SCRAM version of AERMOD and the BREEZE Parallel version of AERMOD. Each AERMOD input file has an .ami file extension. The BREEZE AERMOD Parallel input files have the designator “_MPI” preceding the .ami file extension.
AERMOD output files for both the EPA SCRAM version of AERMOD and the BREEZE Parallel version of AERMOD. Each AERMOD output file has an .aml file extension. The BREEZE AERMOD Parallel output files have the designator “_MPI” preceding the .aml file extension.
AERMOD Parallel Test Cases Single download of the following scenarios.
| AERMOD Parallel Test Cases |
Single download of the following scenarios. |
| AGA |
Flat, Rural, Downwash, Independent |
| Alaska |
Flat, Rural, Downwash, Developmental |
| Baldwin |
Flat, Rural, Non-downwash, Independent |
| Bowline |
Flat, Rural, Downwash, Developmental/Independent* |
| Clifty Creek |
Flat, Rural, Non-downwash, Independent |
| DAEC |
Flat**, Rural, Downwash, Developmental |
| EOCR |
Flat, Rural, Downwash, Independent |
| Indianapolis |
Flat, Urban, Non-downwash, Developmental |
| Kincaid SF6 |
Flat, Rural, Non-downwash, Developmental |
| Kincaid SO2 |
Flat, Rural, Non-downwawh, Developmental |
| Lovett |
Terrain, Rural, Non-downwash, Developmental |
| Martin's Creek |
Terrain, Rural, Non-downwash***, Independent |
| Millstone |
Flat, Rural, Downwash, Developmental |
| Prairie Grass |
Flat, Rural, Non-downwash, Developmental |
| Tracy |
Terrain, Rural, Non-downwash, Independent |
| Westvaco |
Terrain, Rural, Non-downwash, Independent |
| EPA Test Cases |
EPA test cases found on SCRAM’s AERMOD web page |
The BREEZE team is available to answer any remaining questions regarding equivalency between BREEZE AERMOD Parallel / BRMS and the EPA’s SCRAM serial version of AERMOD.
Please contact the BREEZE team at +1-972-661-8881 or breeze@trinityconsultants.com for more information.
AERMOD Version 09292 Equivalency
The BREEZE AERMOD Remote Modeling System executes the U.S. EPA AERMOD Fortran executable that has been parallelized for performance across multiple processors (AERMOD Parallel) by BREEZE developers. No changes to Fortran algorithms have been made. In fact, results generated by the BRMS and BREEZE AERMOD Parallel are identical to those generated by the U.S. EPA AERMOD Fortran executable.
As part of the product development process, BREEZE staff performed hundreds of AERMOD model runs using both the AERMOD Fortran executable available from the U.S. EPA SCRAM Web site (http://www.epa.gov/scram001/dispersion_prefrec.htm) and the BREEZE AERMOD Parallel Fortran executable developed by Trinity Consultants, Inc. The U.S. EPA version number for both of these executables was 09292.
Trinity has made available to all U.S. state and federal regulatory modelers the BREEZE AERMOD Parallel modeling application to assist in their equivalency demonstration. In addition, Trinity has made available, as requested by OAQPS, model runs completed with the SCRAM version of AERMOD and BREEZE AERMOD Parallel. Model scenarios include the test cases available for download from the AERMOD model page at the EPA’s SCRAM website (http://www.epa.gov/scram001/dispersion_prefrec.htm) and those described in a related memo ( http://www.epa.gov/scram001/7thconf/aermod/evalreadme.txt).
An equivalency document is also provided listing the differences between the model output list file generated by the SCRAM version as compared to the BREEZE AERMOD Parallel version. This document demonstrates that the only differences in the two results sets were the time stamps at the top of each output page.
The data files for the test runs can be downloaded using the links provided below.
The zipped data sets identified below contain the following files:
.SFC and .PFL model ready meteorological data files. The .SFC file was modified by changing the version number from 02222 to 06341 to make it compatible with AERMOD (i.e., allow AERMOD to process it without error).
AERMOD input files for both the EPA SCRAM version of AERMOD and the BREEZE Parallel version of AERMOD. Each AERMOD input file has an .ami file extension. The BREEZE AERMOD Parallel input files have the designator “_MPI” preceding the .ami file extension.
AERMOD output files for both the EPA SCRAM version of AERMOD and the BREEZE Parallel version of AERMOD. Each AERMOD output file has an .aml file extension. The BREEZE AERMOD Parallel output files have the designator “_MPI” preceding the .aml file extension.
AERMOD Parallel Test Cases Single download of the following scenarios.
EPA Validation Studies (828 model runs / 424 model comparisons)
The model run comparison report for the following model runs can be viewed here. This is a 7mb file. To save the report to you computer, right mouse click on the link.
|
| AGA |
Flat, Rural, Downwash, Independent |
| Alaska |
Flat, Rural, Downwash, Developmental |
| Baldwin |
Flat, Rural, Non-downwash, Independent |
| Bowline |
Flat, Rural, Downwash, Developmental/Independent* |
| Clifty Creek |
Flat, Rural, Non-downwash, Independent |
| DAEC |
Flat**, Rural, Downwash, Developmental |
| EOCR |
Flat, Rural, Downwash, Independent |
| Indianapolis |
Flat, Urban, Non-downwash, Developmental |
| Kincaid SF6 |
Flat, Rural, Non-downwash, Developmental |
| Kincaid SO2 |
Flat, Rural, Non-downwawh, Developmental |
| Lovett |
Terrain, Rural, Non-downwash, Developmental |
| Martin's Creek |
Terrain, Rural, Non-downwash***, Independent |
| Westvaco |
Terrain, Rural, Non-downwash, Independent |
EPA Test Cases - Standard Set
The Standard test cases provided by EPA are scenarios designed to test 09292 AERMOD options that do not use the non-default FASTALL model option. You can view the model comparison report for each scenario by clicking on the link to the right of the sceanrio name. The only differences observed between the serial EPA version and the BREEZE Parallel version were the date/time stamps of the model runs and some header information.
|
| aertest |
Click here for comparison report |
| allsrcs |
Click here for comparison report |
| capped |
Click here for comparison report |
| capped_nostd |
Click here for comparison report |
| flatelev |
Click here for comparison report |
| hrdow |
Click here for comparison report |
| lovett |
Click here for comparison report |
| mcr |
Click here for comparison report |
| multurb |
Click here for comparison report |
| olm |
Click here for comparison report |
| olmgrp |
Click here for comparison report |
| openpits |
Click here for comparison report |
| psdcred |
Click here for comparison report |
| pvmrm |
Click here for comparison report |
| scimtest |
Click here for comparison report |
| surfcoal |
Click here for comparison report |
| testgas |
Click here for comparison report |
| testgas2 |
Click here for comparison report |
| testpart |
Click here for comparison report |
| testpm10 |
Click here for comparison report |
| testpm10_1986 |
Click here for comparison report |
| testpm10_1987 |
Click here for comparison report |
| testpm10_1988 |
Click here for comparison report |
| testpm10_1989 |
Click here for comparison report |
| testpm10_1990 |
Click here for comparison report |
| testpm25 |
Click here for comparison report |
| testprt2 |
Click here for comparison report |
EPA Test Cases - FASTALL Set
The FASTALL test cases provided by EPA are scenarios designed to test 09292 AERMOD options that do use the non-default FASTALL model option. You can view the model comparison report for each scenario by clicking on the link to the right of the sceanrio name. The only differences observed between the serial EPA version and the BREEZE Parallel version were the date/time stamps of the model runs and some header information
|
| aertest |
Click here for comparison report |
| allsrcs |
Click here for comparison report |
| capped |
Click here for comparison report |
| capped_nostd |
Click here for comparison report |
| flatelev |
Click here for comparison report |
| hrdow |
Click here for comparison report |
| lovett |
Click here for comparison report |
| mcr |
Click here for comparison report |
| multurb |
Click here for comparison report |
| olm |
Click here for comparison report |
| olmgrp |
Click here for comparison report |
| openpits |
Click here for comparison report |
| psdcred |
Click here for comparison report |
| pvmrm |
Click here for comparison report |
| scimtest |
Click here for comparison report |
| surfcoal |
Click here for comparison report |
| testgas |
Click here for comparison report |
| testgas2 |
Click here for comparison report |
| testpart |
Click here for comparison report |
| testpm10 |
Click here for comparison report |
| testpm10_1986 |
Click here for comparison report |
| testpm10_1987 |
Click here for comparison report |
| testpm10_1988 |
Click here for comparison report |
| testpm10_1989 |
Click here for comparison report |
| testpm10_1990 |
Click here for comparison report |
| testpm25 |
Click here for comparison report |
| testprt2 |
Click here for comparison report |
The BREEZE team is available to answer any remaining questions regarding equivalency between BREEZE AERMOD Parallel / BRMS and the EPA’s SCRAM serial version of AERMOD.
Please contact the BREEZE team at +1-972-661-8881 or breeze@trinityconsultants.com for more information.