Differences

This shows you the differences between two versions of the page.

--- abstract:mills:mills [2018-06-07 23:29] – [Compute nodes] anita
+++ abstract:mills:mills [2023-08-21 10:50] (current) – [Getting started on Mills (Retired)] anita
@@ Line 1: / Line 1: @@
-====== Getting started on Mills ======
+====== Getting started on Mills (Retired) ======
 The Mills cluster, UD's first Community Cluster, was deployed in 2012 and is a distributed-memory, Linux cluster. It consists of 200 compute nodes (5160 cores, 14.5 TB memory, 49.3 peak Tflops). The nodes are built of AMD “Interlagos” 12-core processors, in dual- and quad-socket configurations for 24- and 48-cores per node. A QDR InfiniBand network fabric supports the Lustre filesystem (approx 180 TB of usable space). Gigabit and 10-Gigabit Ethernet networks provide access to additional Mills filesystems and the campus network. The cluster was purchased with a proposed 5 year life, putting its retirement in the October 2016 to January 2017 time period.
@@ Line 6: / Line 6: @@
 For general information about the community cluster program, visit the [[itrc>community-cluster|IT Research Computing website]]. To cite the Mills cluster for grants, proposals and publications, use these [[itrc>community-cluster-templates/|HPC templates]].
+Login (head) node is online to allow file transfer off Mills' filesystems.
+<note warning>
+**Monday, August 21,2023**: The Mills cluster was fully retired and is no longer accessible. Any data present in /lustre/work, /archive or /home directories on Mills is no longer available.
+We must finally say goodbye to the Mills cluster. Thank you to all Mills cluster users for your cooperation and contributions to the UD research and HPC communities.
+For complete details see [[https://sites.udel.edu/it-rci/2023/07/31/mills-retirement-on-august-21-2023/|Mills Retirement On August 21, 2023]].
+</note>
 ===== Configuration =====
@@ Line 18: / Line 28: @@
 ==== Login node ====
-The login node is the primary node shared by all cluster users.  Its computing environment is a full standard variant of UNIX configured for scientific applications.  This includes command documentation (man pages), scripting tools, compiler suites, debugging/profiling tools and application software.  In addition, the login node has several tools to help you move files between the HPC filesystem and your local machine, other clusters and web-based services.
+The login node is the primary node shared by all cluster users.  Its computing environment is a full standard variant of UNIX configured for scientific applications.  This includes command documentation (man pages), scripting tools, compiler suites, debugging/profiling tools, and application software.  In addition, the login node has several tools to help you move files between the HPC filesystem and your local machine, other clusters, and web-based services.
 <note warning> Use the login node to set up workflow processes and to compile and debug programs. You should generally use compute nodes to run application software or your own executables.
@@ Line 27: / Line 37: @@
 ==== Compute nodes ====
-There are many compute nodes with different configurations.  Each node may have extra memory, multi-core processors (CPUs), GPUs and/or extra local disk space.  They may have different OS versions or OS configurations, such as mounted network disks.  This document assumes all the compute nodes have the same OS and almost the same configuration.  Some nodes may have more cores, more memory or more disk.
+There are no longer compute nodes.
-The standard UNIX on the compute nodes is configured to support just the running of your jobs, particularly parallel jobs.  For example, there are no man pages on the compute nodes.  Large components of the OS, such as X11, are only added to the environment when needed.
-All the multi-core CPUs and GGPs share the same memory in what may be a complicated manner. To add more processing capability while keeping hardware expense and power requirement down, most architectures use Non-Uniform Memory Access (NUMA).  Also the processors may be sharing hardware, such as the FPUs (Floating point units).
-Commercial applications, and normally your program, will use a layer of abstraction called a //programming model//.  Consult the cluster specific documentation for advanced techniques to take advantage of the low lever architecture.
 ===== Storage =====
@@ Line 43: / Line 46: @@
 ==== High-performance filesystems ====
-One important component of HPC designs is to provide fast access to large files and to many small files.  These days, high-performance filesystems have capacities ranging from hundreds of terabytes to petabytes. They are designed to use parallel I/O techniques to reduce file-access time. The [[:abstract:mills:filesystems:lustre|Lustre]] filesystems, in use at UD, are composed of many physical disks using technologies such as RAID-6 to give resilience, data integrity and parallelism at multiple levels. They use high-bandwidth interconnects such as InfiniBand and 10-Gigabit Ethernet.
+One important component of HPC designs is to provide fast access to large files and to many small files.  These days, high-performance filesystems have capacities ranging from hundreds of terabytes to petabytes. They are designed to use parallel I/O techniques to reduce file-access time. The [[:abstract:mills:filesystems:lustre|Lustre]] filesystems, in use at UD, are composed of many physical disks using technologies such as RAID-6 to give resilience, data integrity, and parallelism at multiple levels. They use high-bandwidth interconnects such as InfiniBand and 10-Gigabit Ethernet.
 High-performance filesystems, such as Lustre, are typically designed as volatile, scratch storage systems. This is because traversing the entire filesystem takes so much time that it becomes financially infeasible to create off-site backups.  However, the RAID-6 design provides increased user-confidence by providing a high level of built-in redundancy against hardware failure.
@@ Line 65: / Line 68: @@
 ===== Help =====
-==== System problems or can't find an answer on this wiki ====
+==== System or account problems, or can't find an answer on this wiki ====
-If you are experiencing a system related problem, first check Mills [[http://mills.hpc.udel.edu/ganglia/|cluster monitoring]] and [[http://www.hpc.udel.edu/mantis/default/project_page.php?project_id=4|system alerts]]. To report a new problem, or you just can't find the help you need on this wiki, then submit a [[http://www.udel.edu/deptforms/it/research_computing/index.html|Research Computing Help Request]] specifying ''High Performance Computing'' and ''Mills cluster'' for the problem details.
+If you are experiencing a system related problem, first check Mills [[http://mills.hpc.udel.edu/ganglia/|cluster monitoring]] and [[http://www.hpc.udel.edu/mantis/default/project_page.php?project_id=4|system alerts]]. To report a new problem, or you just can't find the help you need on this wiki, then submit a [[https://services.udel.edu/TDClient/32/Portal/Requests/TicketRequests/NewForm?ID=D5ZRIgFlfLw_|Research Computing High Performance Computing (HPC) Clusters Help Request]] and complete the form including Mills and your problem details in the description field.
 ==== Ask or tell the HPC community ====
-[[https://groups.google.com/a/udel.edu/d/forum/hpc-ask?hl=en-US|hpc-ask]] is a Google group established to stimulate interactions within UD’s broader HPC community and is based on members helping members. This is a great venue to post a question, start a discussion or share an upcoming event with the community. Anyone may request membership. Messages are sent as a daily summary to all group members. This list is archived, public, and searchable by anyone.
+[[https://groups.google.com/a/udel.edu/d/forum/hpc-ask?hl=en-US|hpc-ask]] is a Google group established to stimulate interactions within UD’s broader HPC community and is based on members helping members. This is a great venue to post a question about HPC, start a discussion, or share an upcoming event with the community. Anyone may request membership. Messages are sent as a daily summary to all group members. This list is archived, public, and searchable by anyone.
 ==== Publication and Grant Writing Resources =====
 [[http://sites.udel.edu/research-computing/community-cluster-templates/|HPC templates]] are available to use for a proposal or publication to acknowledge use of or describe UD’s Information Technologies HPC resources.