AMD EPYC/Threadripper - ZEN 2 Architecture

[ROG-X]

19 ώρες πριν, greg4 είπε

Παιδιά με τέτοια ενθουσιασμό μην ξεχάσετε να πάρετε το μέρισμα από τις μετοχές στην AMD

Α... ξέχασα δεν έχετε μετοχές, κάποιοι άλλοι βγάζουν πολλά $$ όσο εμείς φανατιζόμαστε...

 

Aλλος αρμεγε το γελάδι τοσα χρονια και αλλαζει τα socket σαν τα σωβρακα.

Επεξ/σία 13 Φεβρουαρίου 2020 από ROG-X

[akoinonitos]

1 ώρα πριν, holy grail είπε

ολες real εφαρμογες ειναι.....

μεγαλη επιτυχια παντως να εχει μικροτερο bandwidth απο τον 3960χ και απο τον ταπεινο ιντελ w.

Aμα στους χαριζαν να επαιρνες τον xeon επειδη εχει λιγο μεγαλυτερο read στο aida 🤣

[GoMaR]

Δε ξερω τι λετε..αλλα εσκασε ο 3970x με την οικογενεια του  

[akoinonitos]

2 ώρες πριν, GoMaR είπε

Δε ξερω τι λετε..αλλα εσκασε ο 3970x με την οικογενεια του  

Mήπως σου περισσεύει καμία 2070 

Επεξ/σία 13 Φεβρουαρίου 2020 από akoinonitos

[GoMaR]

πρωτο run χωρις drivers χωρις τιποτα. Ολα στοκ εκτος απο το XMP @3000mhz 

[ilos]

4 ώρες πριν, GoMaR είπε

πρωτο run χωρις drivers χωρις τιποτα. Ολα στοκ εκτος απο το XMP @3000mhz 

Η αμήχανη στιγμή όταν το μεσαίο HEDT part πατάει κάτω ναυαρχίδες XEON.😁

[Aten-Ra]

https://www.tomshardware.com/news/us-navy-supercomputer-cray-amd-nvidia?_ga=2.261088205.1702042986.1581948995-25032607.1539794145

The U.S. Navy Department of Defense Supercomputing Resource Center (DSRC) on Monday announced plans to install its first supercomputer with a peak performance of over 10 petaflops (12.8 petaflopsto be exact). Cray will build the supercomputer on its Shasta platform, and the supercomputer will come packing AMD Epyc CPUs and Nvidia Volta V100 GPUs.

The new Cray supercomputer will feature 290,304 AMD Epyc 7002-series processor cores, 112 Nvidia Volta V100 GPUs, a 200 gigabit per second Cray Slingshot network interconnect, 590TB of memory and 14 petabytes of usable storage. 

 

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------

 

 

https://www.amd.com/en/press-releases/2020-02-18-amd-epyc-cloud-adoption-grows-google-cloud

New Google Compute Engine N2D virtual machines (VMs) based on 2^nd Gen AMD EPYC™ processors provide double digit savings and performance improvements over existing N-series instances

SANTA CLARA, Calif. 

02/18/2020

AMD (NASDAQ: AMD) and Google Cloud today announced the beta availability of N2D VMs on Google Compute Engine powered by 2^nd Gen AMD EPYC^™ processors. The N2D family of VMs is a great option for customers running general purpose and high-performance workloads requiring a balance of compute and memory.

N2D VMs are designed to provide customers with:

• High-performance 2^nd Gen AMD EPYC processors that provide greater flexibility for customers to choose the best VM for their workload and up to a 39% performance improvement on the Coremark benchmark versus comparable N1 instances while offering a savings of up to 13% over comparable N-series instances,
• 128 and 224 vCPUs configuration options that offer up to 70% higher platform memory bandwidth compared to existing comparable VMs in the Google Compute Engine catalogue for HPC workloads requiring high memory bandwidth,
• Access to higher platform memory bandwidth and higher core counts, leading to a 100% performance improvement on a variety of benchmarks, including Gromacs and NAMD, compared to n1-standard-96 vCPUs.

 “Cloud providers and hosters around the world recognize the fantastic core scaling, massive memory bandwidth, impressive TCO savings and record setting performance of the 2^nd Gen AMD EPYC processors,” said Forrest Norrod, senior vice president and general manager, Datacenter and Embedded Solutions Business, AMD. “AMD and Google have worked together closely on these initial VMs to help ensure Google Cloud customers have a high-performance and cost-effective experience across a variety of workloads, and we will continue to work together to provide that experience this year and beyond.”

https://cloud.google.com/blog/products/compute/announcing-the-n2d-vm-family-based-on-amd

 

At Google Cloud, we want you to be able to choose the best VMs for your workloads. Today, we’re excited to announce a new addition to our general purpose VMs: the N2D family, built atop 2nd Gen AMD EPYC™ Processors. 

N2D VMs are a great option for both general-purpose workloads and workloads that require high memory bandwidth.

• General-purpose workloads that require a balance of compute and memory, like web applications and databases, can benefit from N2D’s performance,  price, and features. N2D VMs are designed to provide you with the same features as N2 VMs including local SSD, custom machine types, and transparent maintenance through live migration, while features like large machine types with up to an industry-leading 224 vCPUs, the largest general purpose VM on Compute Engine. At the same time, N2D instances provide savings of up to 13% over comparable N-series instances, and up to a 39% performance improvement on the Coremark benchmark compared to comparable N1 instances¹.
• HPC workloads such as crash analysis, financial modeling, rendering and reservoir analysis, will benefit from the N2D machine types configured with 128 and 224 vCPUs, which offer up to 70% higher platform memory bandwidth than comparable N1 instances. This, combined with higher core counts, provides over a 100% performance improvement on a variety of representative benchmarks, including Gromacs and NAMD, compared to n1-standard-96 vCPUs. 

N2D machine type details

N2D VMs are now available in beta from us-central1, asia-southeast1, and europe-west4, with more regions on the way! You can launch them on-demand or as preemptible VMs. When you sign up for committed use discounts, you can save up to 55% for three-year commitments versus on-demand pricing. Long-running N2D VMs can take advantage of sustained use discounts, and automatically save up to 20%. You can also configure N2D VMs as predefined machine types with vCPU to memory ratios of 1:1, 1:4, and 1:8,  up to 224 vCPUs. You can also create custom machine types with N2Ds, helping you meet the needs of diverse workloads.

Επεξ/σία 19 Φεβρουαρίου 2020 από Aten-Ra

[GoMaR]

λοιπον 
Ανοιξα PBO (δεν εχω ιδεα τι πειραζω) και οι θερμοκρασιες ανεβηκαν αλλα οπως και το score.

Ειδα και εως 18χχχ μπροστα αλλα το συνηθες ειναι 179χχ




Οι θερμοκρασιες μετα απο 20 λουπες ειναι 80C με την enermax TRII 360 και παστα noctua H2

Επισης στα προγραμματα που δουλευω οι θερμοκρασιες ειναι πιο χαμηλες.
 

Επεξ/σία 19 Φεβρουαρίου 2020 από GoMaR

[ROG-X]

Στις 19/2/2020 στις 8:55 ΠΜ, Aten-Ra είπε

https://www.tomshardware.com/news/us-navy-supercomputer-cray-amd-nvidia?_ga=2.261088205.1702042986.1581948995-25032607.1539794145

The U.S. Navy Department of Defense Supercomputing Resource Center (DSRC) on Monday announced plans to install its first supercomputer with a peak performance of over 10 petaflops (12.8 petaflopsto be exact). Cray will build the supercomputer on its Shasta platform, and the supercomputer will come packing AMD Epyc CPUs and Nvidia Volta V100 GPUs.

The new Cray supercomputer will feature 290,304 AMD Epyc 7002-series processor cores, 112 Nvidia Volta V100 GPUs, a 200 gigabit per second Cray Slingshot network interconnect, 590TB of memory and 14 petabytes of usable storage. 

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------

https://www.amd.com/en/press-releases/2020-02-18-amd-epyc-cloud-adoption-grows-google-cloud

New Google Compute Engine N2D virtual machines (VMs) based on 2^nd Gen AMD EPYC™ processors provide double digit savings and performance improvements over existing N-series instances

SANTA CLARA, Calif. 

02/18/2020

AMD (NASDAQ: AMD) and Google Cloud today announced the beta availability of N2D VMs on Google Compute Engine powered by 2^nd Gen AMD EPYC^™ processors. The N2D family of VMs is a great option for customers running general purpose and high-performance workloads requiring a balance of compute and memory.

N2D VMs are designed to provide customers with:

• High-performance 2^nd Gen AMD EPYC processors that provide greater flexibility for customers to choose the best VM for their workload and up to a 39% performance improvement on the Coremark benchmark versus comparable N1 instances while offering a savings of up to 13% over comparable N-series instances,
• 128 and 224 vCPUs configuration options that offer up to 70% higher platform memory bandwidth compared to existing comparable VMs in the Google Compute Engine catalogue for HPC workloads requiring high memory bandwidth,
• Access to higher platform memory bandwidth and higher core counts, leading to a 100% performance improvement on a variety of benchmarks, including Gromacs and NAMD, compared to n1-standard-96 vCPUs.

 “Cloud providers and hosters around the world recognize the fantastic core scaling, massive memory bandwidth, impressive TCO savings and record setting performance of the 2^nd Gen AMD EPYC processors,” said Forrest Norrod, senior vice president and general manager, Datacenter and Embedded Solutions Business, AMD. “AMD and Google have worked together closely on these initial VMs to help ensure Google Cloud customers have a high-performance and cost-effective experience across a variety of workloads, and we will continue to work together to provide that experience this year and beyond.”

https://cloud.google.com/blog/products/compute/announcing-the-n2d-vm-family-based-on-amd

At Google Cloud, we want you to be able to choose the best VMs for your workloads. Today, we’re excited to announce a new addition to our general purpose VMs: the N2D family, built atop 2nd Gen AMD EPYC™ Processors. 

N2D VMs are a great option for both general-purpose workloads and workloads that require high memory bandwidth.

• General-purpose workloads that require a balance of compute and memory, like web applications and databases, can benefit from N2D’s performance,  price, and features. N2D VMs are designed to provide you with the same features as N2 VMs including local SSD, custom machine types, and transparent maintenance through live migration, while features like large machine types with up to an industry-leading 224 vCPUs, the largest general purpose VM on Compute Engine. At the same time, N2D instances provide savings of up to 13% over comparable N-series instances, and up to a 39% performance improvement on the Coremark benchmark compared to comparable N1 instances¹.
• HPC workloads such as crash analysis, financial modeling, rendering and reservoir analysis, will benefit from the N2D machine types configured with 128 and 224 vCPUs, which offer up to 70% higher platform memory bandwidth than comparable N1 instances. This, combined with higher core counts, provides over a 100% performance improvement on a variety of representative benchmarks, including Gromacs and NAMD, compared to n1-standard-96 vCPUs. 

N2D machine type details

N2D VMs are now available in beta from us-central1, asia-southeast1, and europe-west4, with more regions on the way! You can launch them on-demand or as preemptible VMs. When you sign up for committed use discounts, you can save up to 55% for three-year commitments versus on-demand pricing. Long-running N2D VMs can take advantage of sustained use discounts, and automatically save up to 20%. You can also configure N2D VMs as predefined machine types with vCPU to memory ratios of 1:1, 1:4, and 1:8,  up to 224 vCPUs. You can also create custom machine types with N2Ds, helping you meet the needs of diverse workloads.

 Καλα  εκει στο πεντάγωνο δεν ξερουν οτι δεν μπορουν να τα δουλεψουν με Windows 10 ? 😂 

Εδω ολοκληρος Holy Grail, απεφανθη οτι τα AMD ειναι για τα μπαζα...

Holy grail, γελανε και τα τσιμεντα....

Ειπαμε να εισαι hater αλλα εσυ αγορι μου εχεις ξεφύγει...

 

Επεξ/σία 20 Φεβρουαρίου 2020 από ROG-X

[malakudi]

Στις 19/2/2020 στις 12:46 ΜΜ, GoMaR είπε

λοιπον 
Ανοιξα PBO (δεν εχω ιδεα τι πειραζω) και οι θερμοκρασιες ανεβηκαν αλλα οπως και το score.

Ειδα και εως 18χχχ μπροστα αλλα το συνηθες ειναι 179χχ




Οι θερμοκρασιες μετα απο 20 λουπες ειναι 80C με την enermax TRII 360 και παστα noctua H2

Επισης στα προγραμματα που δουλευω οι θερμοκρασιες ειναι πιο χαμηλες.
 

Πήρες την Enermax Liqtech II TR4; Δεν έχεις διαβάσει τι γίνεται με αυτές; Πως και την διάλεξες;

 

[GoMaR]

51 λεπτά πριν, malakudi είπε

Πήρες την Enermax Liqtech II TR4; Δεν έχεις διαβάσει τι γίνεται με αυτές; Πως και την διάλεξες;

Επεσε πολυ μελετη, απο απλη αεροψυκτρα εως custom WC, πολλες αποψεις, πολυ ψαξιμο και τελικα καταληξαμε σε αυτην με το μεγαλο ρισκο.  

[malakudi]

13 λεπτά πριν, GoMaR είπε

Επεσε πολυ μελετη, απο απλη αεροψυκτρα εως custom WC, πολλες αποψεις, πολυ ψαξιμο και τελικα καταληξαμε σε αυτην με το μεγαλο ρισκο.  

Σίγουρα είναι η καλύτερη AiO για TR αλλά εγώ δε θα το ρίσκαρα. Έχε το νου σου στους 6 περίπου μήνες θα αρχίσουν τα προβλήματα με idle temps, αν εξακολουθεί το πρόβλημα στην παραγωγή. Είναι τουλάχιστον πρόσφατα κατασκευασμένη;

Μπορείς να τρέξεις κι ένα blender benchmark και να καταγράψεις max temps;

[GoMaR]

πως καταγραφω max temps ?
 

[narta]

46 λεπτά πριν, GoMaR είπε

Επεσε πολυ μελετη, απο απλη αεροψυκτρα εως custom WC, πολλες αποψεις, πολυ ψαξιμο και τελικα καταληξαμε σε αυτην με το μεγαλο ρισκο.  

Για production environment ειναι απλα ρίσκο. Καλυτερα κατι σε Noctua/be quite και ενα τικ κατω σε ταχυτητες

[GoMaR]

84C βαθμους μαξιμουμ. Ολοι οι πυρηνες κατω απο 4050 δεν επεσαν.
Στο τελευταιο test για καποιο λογο παει υπερβολικα αργα.