address comments from Pareena

content/learning-paths/servers-and-cloud-computing/5g-on-arm/_index.md

@@ -0,0 +1,36 @@
+---
+title: 5G On Arm Handbook
+
+learning_path_main_page: 'yes'
+minutes_to_complete:  10
+
+who_is_this_for: This is a introductory topic for software developers interested in running the 5G stack on Arm
+
+learning_objectives: 
+    - Choose the appropriate Arm server for running the 5G stack
+    - Tune the selected servers for running the 5G stack
+
+prerequisites:
+    - Arm server 
+    - 5G ORAN stack
+
+author_primary: Arm
+
+### Tags
+skilllevels: Introductory
+subjects: 5G
+armips:
+    - Neoverse
+tools_software_languages:
+    - SVE
+    - GCC
+    - 5G ORAN
+operatingsystems:
+    - Linux
+
+### FIXED, DO NOT MODIFY
+# ================================================================================
+weight: 1                       # _index.md always has weight of 1 to order correctly
+layout: "learningpathall"       # All files under learning paths have this same wrapper
+learning_path_main_page: "yes"  # This should be surfaced when looking for related content. Only set for _index.md of learning path content.
+---

content/learning-paths/servers-and-cloud-computing/5g-on-arm/_next-steps.md

@@ -0,0 +1,36 @@
+---
+# ================================================================================
+#       Edit
+# ================================================================================
+
+next_step_guidance: As a next step, we recommend the learning path about getting started with the Arm 5G RAN Acceleration Library.
+
+recommended_path: "/learning-paths/servers-and-cloud-computing/ran/"
+# Link to the next learning path being recommended.
+
+# further_reading links to references related to this path. Can be:
+    # Manuals for a tool / software mentioned   (type: documentation)
+    # Blog about related topics                 (type: blog)
+    # General online references                 (type: website) 
+
+further_reading:
+    - resource:
+        title: 5G Infrastructure
+        link: https://www.arm.com/en/markets/5g/infrastructure
+        type: website
+    - resource:
+        title: Arm RAN Acceleration Library Reference Guide
+        link: https://developer.arm.com/documentation/102249
+        type: documentation
+    - resource:
+        title: 5G RAN for Dummies
+        link: https://www.arm.com/resources/dummies-guide/5g-ran
+        type: documentation
+
+# ================================================================================
+#       FIXED, DO NOT MODIFY
+# ================================================================================
+weight: 21                  # set to always be larger than the content in this path, and one more than 'review'
+title: "Next Steps"         # Always the same
+layout: "learningpathall"   # All files under learning paths have this same wrapper
+---

content/learning-paths/servers-and-cloud-computing/5g-on-arm/_review.md

@@ -0,0 +1,77 @@
+---
+# ================================================================================
+#       Edit
+# ================================================================================
+
+# Always 3 questions. Should try to test the reader's knowledge, and reinforce the key points you want them to remember.
+    # question:         A one sentence question
+    # answers:          The correct answers (from 2-4 answer options only). Should be surrounded by quotes.
+    # correct_answer:   An integer indicating what answer is correct (index starts from 0)
+    # explanation:      A short (1-3 sentence) explanation of why the correct answer is correct. Can add additional context if desired
+
+review:
+    - questions:
+        question: >
+            Which server has better support of Nvidia A100X?
+        answers:
+            - "HPE"
+            - "Foxconn"
+            - "Gigabyte"
+            - "WIWYNN"
+            - "Supermicro"
+            - "All the above"
+        correct_answer: 5
+        explanation: >
+            The Supermicro server is designed to support A100X with correct orientation for its ethernet ports
+    - questions:
+        question: >
+            Which statement is correct?
+        answers:
+            - "Regular Linux kernel is sufficient for supporting the 5G stack"
+            - "Low latency Linux kernel is required for supporting 5G stack"
+        correct_answer: 2                    
+        explanation: >
+            Low Latency Kernel minimizes the time it takes for the operating system to respond to events and processes, it is essential for the latency sensitive processes in the 5G stack.
+    - questions:
+        question: >
+            What is the potential issue with a 2P server?        
+        answers:
+            - "PCIe devices sit on different node from the CPU"
+            - "Cross socket communication overhead"
+            - "Sometimes you can't put the PCIe device and CPU on same node"
+            - "All the above"
+        correct_answer: 4                    
+        explanation: >
+            These are all potential issues related to multiple socket servers.
+    - questions:
+        question: >
+            What is isolcpus?
+        answers:
+            - "Remove the core off access"
+            - "Isolate the core off limit"
+            - "Reserved for real-time or other special purpose tasks"
+        correct_answer: 3                    
+        explanation: >
+            The isolcpus setting is important for dedicated tasks that are not interfered by the kernel
+    - questions:
+        question: >
+            How do you affinitize your program to cores?
+        answers:
+            - "taskset"
+            - "numactl"
+            - "pthread_setaffinity_np"
+            - "All the above"
+        correct_answer: 4                    
+        explanation: >
+            These are all the ways you can assign your program to cpu/cores
+
+
+
+
+# ================================================================================
+#       FIXED, DO NOT MODIFY
+# ================================================================================
+title: "Review"                 # Always the same title
+weight: 20                      # Set to always be larger than the content in this path
+layout: "learningpathall"       # All files under learning paths have this same wrapper
+---

content/learning-paths/servers-and-cloud-computing/5g-on-arm/how-to-1.md

@@ -0,0 +1,56 @@
+---
+title: Choose the appropriate Arm server for running the 5G stack
+weight: 2
+
+### FIXED, DO NOT MODIFY
+layout: learningpathall
+---
+
+## Choose the appropriate Arm 5G Servers
+---
+It is essential to choose the appropriate Arm server to run the 5G stack on. In this section, you can learn about the different configurations of Arm servers the 5G stack has been developed and tested on. You can also learn about what needs to be considered when choosing the appropriate Arm server for this task.
+
+#### Extensive 5G development and testing on Arm servers has been done on the hardware configurations and software components listed below:
+
+     1. 5G-in-one-box (software only)
+
+     2. 5G with Inline L1 Accelerator
+
+     3. 5G with L1 GPU Offload
+
+     4. 5G with Lookaside Accelerator Offload
+
+#### The Arm server platforms listed below have been evaluated to run the 5G stack on:
+
+     1. Gigabyte (Mt. Snow 1P)
+
+     2. WIWYNN (Mt. Jade 2P)
+
+     3. Foxconn (Mt. Collins 2P)
+
+     4. Supermicro (R12SPD 1P)
+
+     5. HPE (ProLiant RL300 Gen11 (P59870-B21) 1P) 
+
+#### Things to consider:
+
+1P (Single Processor) vs 2P (Dual Processors) systems:
+
+For evaluation purposes, a 1P system is recommended which should be sufficient to run any 5G software stacks (L1 physical layer/L2 datalink layer/L3 network layer) considering that the Arm CPU has at least 80 cores. 2P system would be great to run more processes on single box, however, socket communication should be avoided, especially when some PCIe devices sit in a different node across from the CPU. 
+
+1U (One Rack Unit) vs 2U (Two Rack Units) systems:
+
+1U is sufficient to run 5G CN (Core Network) software or pure 5G software stacks while 2U is required when you need additional hardware such as several PCIe cards or a full length PCIe card for L1 processing, for example, the Nvidia A100X and Genvisio.
+
+#### Accommodating PCIe Accelerators
+
+Due to the nature of PCIe devices, you need to carefully select the right Arm server to accommodate the PCIe Accelerators you are going to use.
+
+For full length PCIe cards, you need at least a 2U server, however not every 2U server will support full length/full width PCIe devices. Also, some PCIe full profile devices require more power to run properly, for example, Supermicro's Arm server better accommodate some Inline L1 Accelerator cards.
+
+For PCIe cards taking up two PCIe slots like Nvidia A100X, not every 2U will be right choice. Because A100X combines GPU and Mellanox NIC (Network Interface Card) in one device, you need to make sure its NIC ports face out. At this moment, only Supermicro Arm server has designed its server with the full length PCIe cards such as Nvidia A100X converged card.
+
+
+
+
+