Attachment 4 - 1305M320RNRMN0002 - PWS Sample Task Order One.pdf

ATTACHMENT 4 EARTH PREDICTION INNOVATION CENTER (EPIC)

Department of Commerce (DOC) National Oceanic and Atmospheric Administration (NOAA)

Office of Oceanic and Atmospheric Research (OAR) Office of Weather and Air Quality (OWAQ)

SAMPLE TASK ORDER ONE

PERFORMANCE WORK STATEMENT (PWS)

Earth Prediction Innovation Center (EPIC)

ATTACHMENT 4

1305M320RNRMN0002

Contents

1. INFORMATION 3

2. BACKGROUND 3

3. OBJECTIVES 4

4. SCOPE OF WORK 5

5. TASKS 5

6. DELIVERABLES 6

7. KEY PERSONNEL Error! Bookmark not defined.

8. CONSTRAINTS Error! Bookmark not defined.

9. PERIOD OF PERFORMANCE 8

10. PLACE OF PERFORMANCE 8

11. CONTRACT TYPE 8

12. PERFORMANCE REQUIREMENT SUMMARY (PRS) 8

1. INFORMATION

This Performance Work Statement (PWS) outlines the Earth Prediction Innovation Center (EPIC) requirements for obtaining software engineering and scientific/technical support services on cloud infrastructure and to provide robust user support for easy access to the latest version of the National Oceanic and Atmospheric Administration (NOAA) Unified Forecast System (UFS) which is a community-based, coupled, comprehensive Earth modeling system. The efforts are intended to accelerate community-developed scientific and technological enhancements into the operational applications for Numerical Weather Prediction (NWP).

NOAA is leading the development of the EPIC and the associated scientific work practices that will change the way community integrated earth system modeling is achieved in the United States. EPIC will continuously integrate and develop the community-modeling environment with all partners invested in EPIC through a community-based, coupled, comprehensive Earth modeling system (i.e. UFS) to improve NWP and to accelerate the Research to Operations and Operations to Research (R2O/O2R)R2O/O2R paradigm. The UFS numerical applications span local to global domains and predictive time scales from sub-hourly analyses to seasonal predictions. It is designed to support the Weather Enterprise and to be the source system for NOAA's operational NWP applications.

The Government will rely upon the Contractor to provide EPIC with Scientific and Technical Support Services directly supporting EPIC as outlined in this PWS. The resultant contract task order award will be managed through the Office of Weather and Air Quality (OWAQ) within NOAA’s Oceanic and Atmospheric Research (OAR) Line Office.

2. BACKGROUND

The Department of Commerce (DOC), NOAA, and OAR’s OWAQ supports world-class weather and air quality research that leads to, among other things, advancements to NOAA’s NWP that ultimately saves lives, reduces property damage, and enhances the national economy.

In pursuit of its vision and mission, OWAQ works closely with the National Weather Service (NWS) to help develop and transition to NWS’ operational suite of forecast models weather and air quality research, including hurricanes, severe thunderstorms, heavy precipitation, and air pollution. Additionally, OWAQ utilizes social science to learn how to deal with the uncertainties weather presents and to inform its engagement and communication with researchers, funders, and the public.

Furthermore, OWAQ selects and funds research that supports and fosters collaborations within NOAA's research laboratories and across the weather enterprise (i.e., NOAA, other Federal agencies and entities, state and local governments, academia, and the private sector).

The Weather Research and Forecasting Innovation Act of 2017 (WRFIA, Public Law 115-25) directs NOAA to prioritize improving weather data, modeling, computing, forecasting, and warnings for the protection of life and property and for the enhancement of the national economy. The authorizing language for EPIC, in the National Integrated Drought Information System Reauthorization Act of 2018 (NIDISRA, Public Law 115- 423), calls for NOAA to accelerate community-developed scientific and technological enhancements into the operational applications for NWP, which requires “advancing weather modeling skills, reclaiming and maintaining international leadership in the area of numerical weather prediction, and improving the transition of research into operations.”

The devastating impacts of hazardous events on life, property, and the national economy can be mitigated with accurate forecasts. To meet this need while reclaiming global leadership in weather prediction, NOAA must serve the nation by building a community-modeling environment for innovative, integrated earth system modeling.

EPIC is the catalyst for accurate weather forecasts in the near-term and integrated earth system modeling in the long-term. In the near-term, this means developing NWP skills in the UFS. In the long-term, this means expansion to other operational model applications such as high-resolution convective-allowing models and fully coupled subseasonal to seasonal, water and ocean forecast system. EPIC is working to fulfill its mandate and to exceed customer expectations. For additional information on EPIC, see https://owaq.noaa.gov/Programs/EPIC and on UFS, see https://ufscommunity.org/.

3. OBJECTIVES

The objective of this first task order is to provide NOAA with an assessment and project plan for supporting and building of the UFS applications. The contractor will provide EPIC with affordable and effective Professional Scientific and Technical Support Services that enable EPIC to meet its community mission requirements for making UFS accessible as an open source system with access through GitHub repositories to meet R2O/O2R functions. Additionally, this task order will build the UFS codes and dependent software on Cloud ecosystems using platforms such as AWS, Google, and Azure with a focus on building a system that is interoperable and portable from Cloud to on premise HPC architectures. The contractor will integrate into the UFS governance structure to provide user/customer support for the cloud-based UFS system and HPC platforms relevant for the benefit of the wider EPIC and UFS community.

The primary areas of focus of this task order are described below:

● Provide an overall assessment of the UFS cloud related activities and develop a project plan for comprehensive support of all UFS applications.

● Develop a community-based UFS modeling system that runs on Cloud HPC with robust user support, easy access to the latest version of code, input and output data; a code repository maintained under version control software (Github); thorough, understandable documentation; user-friendly workflow;

software infrastructure; tutorials, workshops, help desk services; and ensuring that developers are involved in user support.

● Communicate broadly EPIC activities that deliver short-term success online and publicly at conferences

The success of this contract is solely dependent on integration of EPIC into the UFS infrastructure and governance to support community-based coupled Earth system model development by providing the community of users with a modern and flexible UFS framework that works on Cloud and can easily transition to conventional HPC owned by UFS partners. It’s imperative to engage the extensive research enterprise in the U.S. for EPIC to succeed. EPIC will create an environment to more efficiently and effectively translate research advances into operational outcomes. Ultimately, EPIC’s support of the UFS will enable NOAA to simplify its production suite of forecasting models from a great many independent systems, each of which has to be improved and maintained, to a single seamless coupled Earth modeling system.

https://owaq.noaa.gov/Programs/EPIC https://ufscommunity.org/

4. SCOPE OF WORK

This Contractor shall:

4.1 Develop a platform-agnostic containerized version of the UFS that can be used on Cloud HPC, meaning there is a documented process for running element(s) of the UFS on Cloud HPC via multiple vendors, and passes the graduate student test (The Graduate Student Test means that a student can easily:

access code, run code, change code, test code for correct operation, and evaluate code with standard diagnostic packages; obtain code documentation, user support, and training; and understand what is needed for their code changes to transition to operations).

4.2 Provide the Weather Enterprise with access to UFS operational model code, user-support services, and NOAA with a method for the secure ingest of community code into NOAA applications (with open source on Github and open data policies based in NOAA’s and other agency best-practices and policies).

4.3 Align software engineers with scientists to: (1) create well-documented cloud-friendly code; and (2) free scientists to perform research. Communicate short-term successes.

4.4 Provide a user-support team with individual accountabilities for answering questions and updating the frequently-asked questions. The primary focus of this service will be to ensure access and usability by all community members, regardless of the level of expertise and nationality.

NOAA’s current capacity to provide operational and research-related High-Performance Computing is a limitation to the advancement of numerical weather prediction. The transition of research and development to a cloud environment has the potential to dramatically improve NOAA’s capacity. It is NOAA's intention with EPIC to fully test the limits of this possibility, consequently, the plan that is developed under this RFP shall incorporate this.

5. TASKS

5.1 : Assessment of R2O/O2R of the Current and Future Unified Forecast System

The contractor shall:

5.1.1 Describe your vision for engaging the Weather Enterprise (i.e., UFS users contributing code for potential operational use within NOAA and methods to grow user base) to meet the mandates of the

WRFIA.

5.1.2 Define, develop, and execute a clear project plan with deliverables and success metrics that outline the path to supporting the UFS

5.1.3 Recruit and establish a team of professional software, systems, infrastructure, data scientists, and computational engineers to work directly with physical scientists in an integrated, semi-virtual, cross-disciplinary environment to ensure that the UFS code is supported, easily maintainable, extensible and performance optimized.

5.1.4 Describe and define the optimal mechanisms for supporting and accelerating the R2O/O2R process through this contract.

5.1.5 Align with science goals of the UFS Applications.

5.1.6 Assess and provide a comprehensive report on UFS cloud activities to avoid duplication of effort.

5.1.7 Work with parties to define the vetting process of UFS code back into the master repository.

5.1.8 Assess the UFS Repository Management Plan and the UFS Research to Operations Plan to provide feedback to the UFS community.

http://ufs-dev.rap.ucar.edu/index.html#/science/gst

5.1.9 Survey, assess, and provide a report on current UFS related activities within NOAA utilizing cloud services such as continuous integration (CI), continuous development (CD), data services, debugging, model integration, reforecasts/reanalysis, and retrospectives to avoid duplication of effort and identify gaps.

5.1.10 Provide advice and best practices to ensure a UFS code that is innovative, modern, maintainable, extensible, and performance optimized on modern computer architectures. Include best practices on CI/CD approaches that automatically build and test all proposed code changes in its testing infrastructure.

5.1.11 Work with relevant NOAA labs and centers (i.e., EMC) to develop CI/CD process.

5.1.12 Ensure the hierarchical testing framework works across all UFS modeling components.

5.1. 13 Provide a description of how to design and implement an enterprise-level agile approach to accelerate software testing and integration of the UFS through secure ingest procedures and accelerating the R2O/O2R process.

5.2: Develop and document a cloud-based Unified Forecast System

The contractor shall:

5.2.1 Establish access of the cloud-based Unified Forecast System (UFS), source codes, and dependent libraries (or other open source services) on Github repositories for the EPIC community users.

5.2.2 Populate necessary dependent data through NOAA provided services and infrastructure.

5.2.3 Distribute a workflow (i.e., CIME) for cloud-based UFS on Github repositories.

5.2.4 Port and assess the Weather Application Unified Forecast System (UFS) on Cloud Platforms (e.g., AWS, Google and Azure) fitness for use.

5.2.5 Provide container (e.g., docker, singularity, charliecloud, kubernetes) version of the UFS Weather Application and provide and support container technology to improve developer productivity, efficiency, and application portability.

5.2.6 Port UFS to cloud services for the purpose of UFS activities such as continuous integration, data services, model integration, reforecasts/reanalysis, and ensuring portability.

5.2.7 Work with provided communications platform(s) for interaction within the UFS community that includes researchers, developers and users from NOAA, educational institutions, other federal agencies, and the private sector.

6. DELIVERABLES

Deliverable Frequency Medium/Format/# of Copies

Submit To

Provide an overall assessment of the UFS cloud related activities and develop a project plan for comprehensive support of all UFS applications

Within 3-6 months of award

MS Word Report delivered by way of email

EPIC Program Manager PM and COR

Develop and Document a Containerized Cloud-based UFS Weather Application

Within 6 months of award

Online through UFS Portal and UFS Github Repository

Documentation online

Kick Off Meeting Within 10 business days of award

COR/TM

Monthly Activity Report

By the 15th of every month or with invoice?

via email COR/TM

Weekly Status Report Once a week Meeting with PM or otherwise specified

COR/TM

Level of Effort Spend Plan Within 10 business days of award; ongoing updated on a monthly basis by the 15th of the month via email COR/TM

Task 1: Deliverables

6.1.1 Conduct overall assessment of cloud readiness on the current UFS applications from the existing level of effort. Assess which pieces of the UFS are ready for the cloud on three platforms (AWS, Google and Azure).

6.1.2 Assess performance of container vs. non-container version of the current release of the UFS.6.1.3 Assess cloud readiness of Common Infrastructure for Modeling the Earth (CIME) workflows in three cloud environments (i.e., AWS, Google and Azure).

6.1.4 Port the UFS system on three primary (AWS, Google and Azure) cloud platforms and benchmark the performance on all platforms for UFS releases.

6.1.5 Create report on EPIC’s UFS project plan.

6.1.6 Review UFS open source codes for understanding and potential gaps for cloud HPC performance.

Task 2: Deliverables

6.2.1 Provide technical and user support for cloud-based UFS and support transition of community-based UFS code to operations through the R2O/O2R process.

6.2.2 Define the community of users and define the pathways into the UFS repositories with the UFS governance structure.

6.2.3 Provide a dashboard for users of UFS

6.2.4 Develop and maintain online documentation of the UFS as living documents.

6.2.5 Generate the inventory of the transitions of community code into NOAA applications (count; lines of code).

6.2.6 Provide user support services through the USF community online forum established by NOAA.

6.2.7 Provide documentation functions within collaborative GitHub repositories, web-based documentation, user’s guides, and developers guides (e.g., a wiki).

6.2.8 Generate tutorials as short videos (YouTube) and consistent webinars (Google Hangouts) on how-to run UFS on three cloud platforms (AWS, Google and Azure).

6.2.9 Connect scientists with software engineers through online tutorials.

6.2.10 Conduct workshops as routine in-person user-focused workshops and sessions/fora at meetings

(e.g., AMS and AGU).

6.2.11 Perform code sprints and hack-a-thons as opportunities to work on bounded problem areas;

build partnerships or even pair with other developers; and build camaraderie around the projects.

6.2.12 Participate in the design and implementation of up to 8 meetings for strategic planning, scientific workshops, tutorials, and/or hackathons related to the UFS.

7. PERIOD OF PERFORMANCE

The Task Order Period of Performance (POP) will be one (1) year from date of award with one (1), one-year option period.

8. PLACE OF PERFORMANCE

The place of performance is to be determined (TBD) and shall be proposed by the contractor. Remote work shall be authorized for this requirement.

9. CONTRACT TYPE

This is a firm-fixed price (FFP) order.

10. PERFORMANCE REQUIREMENT SUMMARY (PRS)

Required Service PWS Section Performance Standard Method(s) of

Surveillance Incentive

In alignment with the vision for EPIC, develop a project plan that enables a cross-disciplinary team to work together to ensure that the UFS code is supported, easily maintainable, extensible and performance

3., 5.1.1, 5.1.2, 5.1.3, 6.1.1, and 6.1.5

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Surveillance Incentive optimized; and recruit and develop a team of professional software, systems, infrastructure, data scientists, and computational engineers who will implement the plan Develop a platform-agnostic containerized version of the UFS that can be used on Cloud HPC, meaning there is a documented process for running element(s) of the UFS on Cloud HPC via multiple vendors, and passes the graduate student test.

PWS 4.1 Adherence to software design requirements;

Functionality meets or exceeds NOAA’s and the weather community’s business and technical needs;

Design accommodates ongoing software revisions and extensions with minimal effort;

Documentation is complete, accurate, and understandable for use by the weather community Adherence to schedule

Review of design specification document (if applicable) by Program Manager or designee(s);

Potentially independent testing and verification of Contractor software by NOAA technical staff;

Periodic inspection;

Random monitoring;

Customer feedback;

Consultation with Program Manager, members of the weather community, and other subject matter experts

Possible Inclusion in Past Performance Review

Provide the Weather Enterprise with access to UFS operational model code, user-support services, and NOAA with a method for the secure ingest of community code into NOAA applications

PWS 4.2 Adherence to software requirements, design specification, and development methodology (with open source on Github and open data policies based in NOAA’s and other agency best-practices and policies);

Adherence to schedule

Potentially independent testing and verification of software by NOAA technical staff;

Approval by NOAA designated Change Management Board for deployment;

Random inspections and evaluation by NOAA technical staff/SME’s, under direction of Program Manager, COR, or both

Possible Inclusion in Past Performance Review

Align software engineers with scientists to: (1) create well-documented cloud-friendly code;

and (2) free scientists to perform research.

Communicate short-term successes.

PWS 4.3 Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs;

Adherence to schedule

Random inspections;

Customer feedback;

Consultation with Program Manager, members of the weather community, and other subject matter experts

Possible Inclusion in Past Performance

Surveillance Incentive

Provide a user-support team with individual accountabilities for answering questions and updating the frequently-asked questions.

PWS 4.4 Access and usability by all community members, regardless of the level of expertise and nationality

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Describe and define the optimal mechanisms for supporting and accelerating the R2O/O2R process through this contract

PWS

5.1.4

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Align with science goals of the UFS Applications

PWS

5.1.5

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Assess and provide a comprehensive report on UFS cloud activities to avoid duplication of effort

PWS

5.1.6

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Work with parties to define the vetting process of UFS code back into the master repository

PWS

5.1.7

Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs;

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Assess the UFS Repository Management Plan and the UFS Research to Operations Plan to provide feedback to the UFS community

PWS

5.1.8

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Survey, assess, and provide a report on current UFS related activities within NOAA utilizing cloud services such as continuous integration (CI), continuous

PWS

5.1.9

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past

Surveillance Incentive development (CD), data services, debugging, model integration, reforecasts/reanalysis, and retrospectives to avoid duplication of effort and identify gaps Provide advice and best practices to ensure a UFS code that is innovative, modern, maintainable, extensible, and performance optimized on modern computer architectures. Include best practices on CI/CD approaches that automatically build and test all proposed code changes in its testing infrastructure

PWS

5.1.10

Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs;

Adherence to schedule – due within 3-6 months of award

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Work with relevant NOAA labs and centers (i.e., EMC) to develop CI/CD process

PWS

5.1.11

Adherence to software requirements, design specification, and development methodology

Consultation with relevant NOAA technical staff

Possible Inclusion in Past Performance Review

Ensure the hierarchical testing framework works across all UFS modeling components

PWS

5.1.12

Adherence to software requirements, design specification, and development methodology;

Functionality of framework meets or exceeds NOAA’s and the weather community’s needs

Review of design specification document (if applicable) by PM or designee(s);

Potentially independent testing and verification of Contractor software by NOAA technical staff;

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Provide a description of how to design and implement an enterprise-level agile approach to accelerate software testing and integration of the UFS through secure ingest procedures and accelerating the R2O/O2R process

PWS

5.1.13

Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs;

100% inspection by the PM and COR as well as consultation with relevant NOAA technical staff

Possible Inclusion in Past

Surveillance Incentive

Adherence to schedule – due within 3-6 months of award

Establish access of the cloud-based Unified Forecast System (UFS), source codes, and dependent libraries (or other open source services) on Github repositories for the EPIC community users

PWS

5.2.1

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Populate necessary dependent data through NOAA provided services and infrastructure

PWS

5.2.2

Complete, accurate, reliable and functional data management, datasets, and application operation

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Distribute a workflow (i.e., CIME) for cloud-based UFS on Github repositories

PWS

5.2.3

Functionality of workflow meets or exceeds NOAA’s and the weather community’s needs

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Port and assess the Weather Application Unified Forecast System (UFS) on Cloud Platforms (e.g., AWS, Google and Azure) fitness for use

PWS

5.2.4

Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Provide container version of the UFS Weather Application and provide and support container technology to improve developer productivity, efficiency, and application portability

PWS

5.2.5

Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and the weather community’s needs

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past Performance Review

Port UFS to cloud services for the purpose of UFS activities such as continuous integration, data services, model integration, PWS

5.2.6

Adherence to software requirements, design specification, and development methodology;

Functionality of containerized code meets or exceeds NOAA’s and

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past

Surveillance Incentive reforecasts/reanalysis, and ensuring portability the weather community’s needs

Work with provided communications platform(s) for interaction within the UFS community that includes researchers, developers and users from NOAA, educational institutions, other federal agencies, and the private sector

PWS

5.2.7

Timely, relevant, effective, and responsive interactions with the UFS community

Periodic inspection;

Random monitoring;

Customer feedback

Possible Inclusion in Past