diff --git a/docs/get-started/cost-billing/billing-faqs.md b/docs/get-started/cost-billing/billing-faqs.md new file mode 100644 index 00000000..55b4f953 --- /dev/null +++ b/docs/get-started/cost-billing/billing-faqs.md @@ -0,0 +1,104 @@ +# Billing Frequently Asked Questions (FAQs) + +Our primary focus is to deliver outstanding on-prem cloud services, prioritizing +reliability, security, and cutting-edge solutions to meet your research and teaching +requirements. To achieve this, we have implemented a cost-effective pricing model +that enables us to maintain, enhance, and sustain the quality of our services. By +adopting consistent cost structures across all institutions, we can make strategic +investments in infrastructure, expand our service portfolio, and enhance our +support capabilities for a seamless user experience. + +Most of the institutions using our services have an **MOU (Memorandum Of Understanding)** +with us to be better aligned to a number of research regulations, policies and +requirements but if your institution does not have an MOU with us, please have +someone from your faculty or administration contact us to discuss it soon by emailing +us at [help@nerc.mghpcc.org](mailto:help@nerc.mghpcc.org?subject=NERC%20MOU%20Question) +or, by submitting a new ticket at [the NERC's Support Ticketing System (osTicket)](https://mghpcc.supportsystem.com/open.php). + +## Questions & Answers + +???+ question "1. How often will I be billed?" + + You or your institution will be billed monthly within the first week of each + month. + +??? question "2. If I have an issue with my bill, who do I contact?" + + Please send your requests by emailing us at + [help@nerc.mghpcc.org](mailto:help@nerc.mghpcc.org?subject=NERC%20MOU%20Question) + or, by submitting a new ticket at [the NERC's Support Ticketing System (osTicket)](https://mghpcc.supportsystem.com/open.php). + +??? question "3. How do I control costs?" + + Upon creating a project, you will set these resource limits (quotas) for + OpenStack (VMs), OpenShift (containers), and storage through [ColdFront](../get-an-allocation.md). + This is the maximum amount of resources you can consume at one time. + +??? question "4. Are we invoicing for CPUs/GPUs only when the VM or Pod is active?" + + Yes. You will only be billed based on your utilization (cores, memory, GPU) + while your VM or Pod is on. Utilization will be translated into billable + [Service Units (SUs)](how-pricing-works.md#service-units-sus). Persistent + storage related to an OpenStack VM or OpenShift Pod will continue to be + billed even when the VM or Pod is off. + +??? question "5. Will OpenStack & OpenShift show on a single invoice?" + + Yes. In the near future customers of NERC will be able to view per project service + utilization via the XDMoD tool. + +??? question "6. What happens when a Flavor is expanded during the month?" + + a. Flavors cannot be expanded. + + b. You can create a snapshot of an existing VM/Instance and, with that snapshot, + deploy a new flavor of VM/Instance. + +??? question "7. Is storage charged separately?" + + Yes, but on the same invoice. To learn more, see [our page on Storage](how-pricing-works.md#storage). + +??? question "8. Will I be charged for storage attached to shut-off instances?" + + Yes. + +??? question "9. Are we Invoicing Storage using ColdFront Requests or resource usage?" + + a. Storage is invoiced based on [Coldfront Requests](../get-an-allocation.md#how-to-request-a-new-resource-allocation). + + b. When you request additional storage through Coldfront, invoicing on that + additional storage will occur when your request is fulfilled. When you request + a decrease in storage through [Request change using ColdFront](../get-an-allocation.md#request-change-to-resource-allocation-to-an-existing-project), + your invoicing will adjust accordingly when your request is made. In both cases + 'invoicing' means 'accumulate hours for whatever storage quantity was added + or removed'. + + For example: + + 1. I request an increase in storage, the request is approved and processed. + - At this point we start Invoicing. + + 2. I request a decrease in storage. + - The invoicing for that storage stops immediately. + +??? question "10. For OpenShift, what values are we using to track CPU & Memory?" + + a. For invoicing we utilize `requests.cpu` for tracking CPU utilization & + `requests.memory` for tracking memory utilization. + + b. Utilization will be capped based on the limits you set in ColdFront for + your resource allocations. + +??? question "11. If a single Pod exceeds the resources for a GPU SU, how is it invoiced?" + + It will be invoiced as 2 or more GPU SU's depending on how many multiples of + the resources it exceeds. + +??? question "12. How often will we change the pricing?" + + a. Our current plan is no more than once a year for existing offerings. + + b. Additional offerings may be added throughout the year (i.e. new types of + hardware or storage). + +--- diff --git a/docs/get-started/cost-billing/billing-process-for-bu.md b/docs/get-started/cost-billing/billing-process-for-bu.md new file mode 100644 index 00000000..df0ebf1d --- /dev/null +++ b/docs/get-started/cost-billing/billing-process-for-bu.md @@ -0,0 +1,22 @@ +# Billing Process for Boston University + +Boston University has elected to receive a centralized invoice for its university +investigators and their designated user’s use of NERC services. IS&T will then +internally recover the cost from investigators. The process for cost recovery is +currently being implemented, and we will reach out to investigators once the process +is complete to obtain internal funding information to process your monthly bill. + +## Subsidization of Boston University’s Use of NERC + +Boston University will subsidize a portion of NERC usage by its investigators. +The University will subsidize $100 per month of an investigator’s total usage on +NERC, regardless of the number of NERC projects an investigator has established. +Monthly subsidies cannot be carried over to subsequent months. The subsidized +amount and method are subject to change, and any adjustments will be conveyed +directly to investigators and updated on this page. + +Please direct any questions about BU’s billing process by emailing us at +[help@nerc.mghpcc.org](mailto:help@nerc.mghpcc.org?subject=NERC%20Boston%20University%20Billing%20Question) +or submitting a new ticket to the [the NERC's Support Ticketing System (osTicket)](https://mghpcc.supportsystem.com/open.php). + +--- diff --git a/docs/get-started/cost-billing/billing-process-for-harvard.md b/docs/get-started/cost-billing/billing-process-for-harvard.md new file mode 100644 index 00000000..857c9f91 --- /dev/null +++ b/docs/get-started/cost-billing/billing-process-for-harvard.md @@ -0,0 +1,58 @@ +# Billing Process for Harvard University + +NERC billing at Harvard will be similar to how AWS accounts are billed and leverage +the HUIT Central Billing [Portal](https://billing.huit.harvard.edu/). Your HUIT +Customer Code will be matched to your NERC Project Allocation Name as a Billing +Asset. In this process you will be asked for your GL billing code, which you can +change as needed per project. Please be cognizant that only a single billing code +is allowed per billing asset. Therefore, if you have multiple projects with different +funds, if you are able, please create a separate project for each fund. Otherwise, +you will need to take care of this with internal journals inside of your department +or lab. Each monthly billing cycle, the HUIT billing team will review and update +the cost per asset. Subsequently, they will email billing cost information for +each asset to the corresponding billing email addresses of all customers. + +Please follow these two steps to ensure proper billing setup: + +1. Each Harvard PI must have a HUIT billing account linked to their NetID (abc123). +NERC must have HUIT “Customer Code” for billing purposes. To create a HUIT billing +account, sign up [here](https://billing.huit.harvard.edu/portal/allusers/newcustomer) +with your HarvardKey. + + !!! abstract "What if you already have an existing Customer Code?" + *Please note that if you already have an existing active NERC account, you + need to provide your HUIT Customer Code to NERC. If you think your department + may already have a HUIT account but you don’t know the corresponding Customer + Code then you can [contact HUIT Billing](https://billing.huit.harvard.edu/portal/allusers/contactus) + to get the required Customer Code.* + +2. Provide NERC your six-digit HUIT “Customer Code” via +[help@nerc.mghpcc.org](mailto:help@nerc.mghpcc.org?subject=NERC%20HUIT%20Customer%20Code%Details) +and specify which OpenShift or OpenStack project this is for. + + !!! info "How to get the Allocated Project Name and Allocated Project ID?" + By clicking on the Allocation detail page through ColdFront, you can access + information about the allocation of each resource, including OpenStack and + OpenShift as described [here](../get-an-allocation.md#general-user-view). + You will need to provide us **Allocated Project Name** and **Allocated Project + ID** attributes, which are located under the “Allocation Attributes” + section on the detail page. + + Once NERC has receive the six-digit HUIT Customer Code for the PI, the NERC + admin team will create a new ServiceNow ticket by contacting + [HUIT Billing](https://billing.huit.harvard.edu/portal/allusers/contactus) + or by emailing HUIT Billing directly at + [huit-billing@harvard.edu](mailto:huit-billing@harvard.edu?subject=HUIT%20Customer%20Code%For%20NERC) + for all approved and active allocation requests. In this email, NERC admin + needs to specify the Customer Code and Unique Project Allocation Name. This + will be mapped to a unique Asset ID in the HUIT billing portal. + + !!! danger "Harvard University Security Policy Information" + *Please note that all assets deployed to your NERC project must be compliant + with University Security policies. Please familiarize yourself with the + Harvard University Information Security Policy and your role in securing + data: [https://policy.security.harvard.edu/](https://policy.security.harvard.edu/). + If you have any questions about how Security should be implemented in the + Cloud, please contact your school security officer: [https://security.harvard.edu/](https://security.harvard.edu/).* + +--- diff --git a/docs/get-started/cost-billing/billing-process-for-my-institution.md b/docs/get-started/cost-billing/billing-process-for-my-institution.md new file mode 100644 index 00000000..b9d9bd65 --- /dev/null +++ b/docs/get-started/cost-billing/billing-process-for-my-institution.md @@ -0,0 +1,42 @@ +# Billing Process for My Institution + +## Memorandum of Understanding (MOU) + +The New England Research Cloud (NERC) is a shared service offered through the +Massachusetts Green High Performance Computing Center (MGHPCC). The MGHPCC will +enter into a lightweight Memorandum of Understanding (MOU) with each institutional +customer that consumes NERC services. The MOU is intended to ensure the institution +maintains access to valuable and relevant cloud services provided by the MGHPCC +via the NERC to be better aligned to a number of research regulations, policies, +and requirements and also ensure NERC remains sustainable over time. + +## Institutions with established MOUs and Billing Processes + +For cost recovery purposes, institutional customers may elect to receive one invoice +for the usage of NERC services by its PIs and cost recovery internally. Every month, +the NERC team will export, back up, and securely store the billing data for all +PIs in the form of comma-separated values (CSV) files and provide it to the MGHPCC +for billing purposes. + +The following institutions using our services have established MOU as well as +billing processes with us: + +1. [Harvard University](billing-process-for-harvard.md) + +2. [Boston University](billing-process-for-bu.md) + +!!! info "Upcoming MOU with other institutions" + We are in the process of establishing MOUs for institutions not listed above. + +PIs from other institutions not listed above can still utilize NERC services with +the understanding that they are directly accountable for managing their usage and +ensuring all service charges are paid promptly. If you have any some common +questions or need further information, see our [Billing FAQs](billing-faqs.md) +for comprehensive answers. + +If your institution does not have an MOU with us, please have someone from your +faculty or administration contact us to discuss it soon by emailing us at +[help@nerc.mghpcc.org](mailto:help@nerc.mghpcc.org?subject=NERC%20MOU%20Question) +or, by submitting a new ticket at [the NERC's Support Ticketing System (osTicket)](https://mghpcc.supportsystem.com/open.php). + +--- diff --git a/docs/get-started/cost-billing/how-pricing-works.md b/docs/get-started/cost-billing/how-pricing-works.md new file mode 100644 index 00000000..3c12f393 --- /dev/null +++ b/docs/get-started/cost-billing/how-pricing-works.md @@ -0,0 +1,127 @@ +# How does NERC pricing work? + +NERC offers you a **pay-as-you-go** approach for pricing for our cloud infrastructure +offerings (Tiers of Service), including Infrastructure-as-a-Service (IaaS) – Red +Hat OpenStack, Platform-as-a-Service (PaaS) – Red Hat OpenShift and NERC Storage +Tiers. NERC offers a flexible cost model where an institution (with a per-project +breakdown) is billed solely for the duration of the specific services required. +Access is based on project-approved resource quotas, eliminating runaway usage and +charges. There are no obligations of long-term contracts or complicated licensing +agreements. Each institution will enter a lightweight MOU with MGHPCC that defines +the services and billing model. + +## Calculations + +### Service Units (SUs) + +![Service Units (SUs)](images/su.png) + +## Breakdown + +### CPU/GPU SUs + +Service Units (SUs) can only be purchased as a whole unit. We will charge for Pods +and VMs on a per-hour basis for any portion of an hour they are used, and any VM +"flavor"/Pod reservation is charged as a multiplier of the base SU for the maximum +resource they reserve. + +**GPU SU Example:** + +- A Pod or VM with: + + `1 A100 GPU, 24 vCPUs, 95MB RAM, 199.2hrs` + +- Will be charged: + + `1 A100 GPU SUs x 200hrs (199.2 rounded up) x $1.803` + + `$360.60` + +**CPU Example:** + +- A Pod or VM with: + + `3 vCPU, 20 GB RAM, 720hrs (24hr x 30days)` + +- Will be charged: + + `5 CPU SUs due to the extra RAM (20GB vs. 12GB(3 x 4GB)) x 720hrs x $0.013` + + `$46.80` + +OpenShift Pods are summed up to the project level so that fractions of CPU/RAM +that some pods use will not get overcharged. There will be a split between CPU and +GPU pods, as GPU pods cannot currently share resources with CPU pods. + +### Storage + +Storage is charged separately at a rate of $0.009 TB/hr or $9.00E-12 KB/hr at a +granularity of KB/hr. For ease of reporting, storage SUs are shown in GB but calculated +in KB. OpenStack volumes remain provisioned until they are deleted. VM’s reserve +volumes, and you can also create extra volumes yourself. In OpenShift pods, storage +is only provisioned while it is active, and in persistent volumes, storage remains +provisioned until it is deleted. + +**Storage Example 1:** + +- Volume or VM with: + + `500GB for 699.2hrs` + +- Will be charged: + + `.5 Storage TB SU (.5 TB x 700hrs) x $0.009 TB/hr` + + `$3.15` + +**Storage Example 2:** + +- Volume or VM with: + + `10TB for 720hrs (24hr x 30days)` + +- Will be charged: + + `7,200 Storage TB SU (10TB x 720 hrs) x $0.009 TB/hr` + + `$648` + +Storage includes all types of storage Object, Block, Ephemeral & Image. + +!!! note "Please note" + After issuing the *shutoff* command to an OpenStack VM, it will not use CPU + or GPU but will continue to use storage. Any extra *Images* you create will + also use storage. + +### High Level Function + +To provide a more practical way to calculate your usage, here is a function of +how the calculation works for OpenShift and OpenStack. + +1. **OpenStack** = (Resource (vCPU/RAM/vGPU) assigned to VM flavor converted to +number of equivalent SUs) * (time VM has been running), rounded up to a whole +hour + Extra storage. + + !!! info "NERC's OpenStack Flavor List" + You can find the most up-to-date information on the current NERC's OpenStack + flavors with corresponding SUs by referring to [this page](../../openstack/create-and-connect-to-the-VM/flavors.md). + +2. **OpenShift** = (Resource (vCPU/RAM/vGPU) requested by Pod converted to number +of equivalent SUs) * (time Pod was running), rounded up to a whole hour. + +## How to Pay? + +To ensure a comprehensive understanding of the billing process and payment options +for NERC offerings, we advise PIs/Managers to [visit individual pages designated +for each institution](billing-process-for-my-institution.md). These pages provide +detailed information specific to each organization's policies and procedures +regarding their billing. By exploring these dedicated pages, you can gain insights +into the preferred payment methods, invoicing cycles, breakdowns of cost components, +and any available discounts or offers. Understanding the institution's unique +approach to billing ensures accurate planning, effective financial management, +and a transparent relationship with us. + +If you have any some common questions or need further information, see our +[Billing FAQs](billing-faqs.md) for comprehensive answers. + +--- diff --git a/docs/get-started/cost-billing/images/su.png b/docs/get-started/cost-billing/images/su.png new file mode 100644 index 00000000..452ac02c Binary files /dev/null and b/docs/get-started/cost-billing/images/su.png differ diff --git a/docs/get-started/images/coldfront-openshift-allocation-general-user-view.png b/docs/get-started/images/coldfront-openshift-allocation-general-user-view.png index f60f587f..cbafa1a9 100644 Binary files a/docs/get-started/images/coldfront-openshift-allocation-general-user-view.png and b/docs/get-started/images/coldfront-openshift-allocation-general-user-view.png differ diff --git a/docs/get-started/images/coldfront-openstack-allocation-general-user-view.png b/docs/get-started/images/coldfront-openstack-allocation-general-user-view.png index b324610e..c442f881 100644 Binary files a/docs/get-started/images/coldfront-openstack-allocation-general-user-view.png and b/docs/get-started/images/coldfront-openstack-allocation-general-user-view.png differ diff --git a/mkdocs.yml b/mkdocs.yml index 2db14e2a..a49e3f7f 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -11,6 +11,13 @@ nav: - Home: index.md - About: about.md - Getting Started: + - Cost & Billing: + - How does NERC pricing work?: get-started/cost-billing/how-pricing-works.md + - Billing Process: + - Billing Process for My Institution: get-started/cost-billing/billing-process-for-my-institution.md + - Billing Process for Harvard University: get-started/cost-billing/billing-process-for-harvard.md + - Billing Process for Boston University: get-started/cost-billing/billing-process-for-bu.md + - Billing FAQs: get-started/cost-billing/billing-faqs.md - User Onboarding Process: get-started/user-onboarding-on-NERC.md - How to Create a User Account: get-started/create-a-user-portal-account.md - How to Get An Allocation: get-started/get-an-allocation.md @@ -126,4 +133,6 @@ extra: # Extensions markdown_extensions: - admonition + - pymdownx.details + - pymdownx.superfences - tables diff --git a/nerc-theme/main.html b/nerc-theme/main.html index 265ee4ee..556d67da 100644 --- a/nerc-theme/main.html +++ b/nerc-theme/main.html @@ -2,14 +2,15 @@ {% block announce %}