All Activity

This stream auto-updates     

  1. Today
  2. velketor

    850 EVO vs 970 EVO

    Source and destination of each 8GB test were the same. I just did a copy/paste inside the same folder. For example, 8GB file in 1 folder on 850 Evo Drive I just copy/pasted right next to itself on the 850 Evo Drive. I then did the same, but for the 970 drive. Saw very different times for transfers. CrystalDiskMark is showing the 970 to be much faster. Seems like every benchmark tool has a completely different result. Is CrystalDiskMark a trusted source, since Samsung Magician appears to be wildly inaccurate? 970 Evo: 850 Evo:
  3. Hi, I'm trying to upgrade my Brand New Xbox One X to a 2TB SSD. I don't know TOO much about SSD's but I've created a final list of the ones I am choosing between to buy. I was wondering if anyone can give me their opinions on which one I should go with. Here's the list: Seagate Barracuda SATA III 2TB SSD: ADATA SU800 SATA III 2TB SSD: SanDisk Ultra 2TB SATA III SSD: Samsung 860 QVO 2TB SATA III SSD: Lastly, The Samsung 860 EVO which I am including in to compare it to the others. While it is a little out of my budget but I may have to go with it if the others are bad. I really appreciate any help.
  4. Yesterday
  5. To address this, AWS released the new G4 instances that leverage the latest generation NVIDIA T4 GPUs, custom 2nd Generation Intel Xeon Scalable (Cascade Lake) processors, up to 100 Gbps of networking throughput, and up to 1.8TB of local NVMe storage, to deliver the most cost-effective GPU instances for machine learning inference. The new G4 instances can deliver up to 65 TFLOPs of mixed-precision performance, great for inferencing. G4’s can also be used on cost-effectively for small-scale and entry-level machine learning training jobs that are less sensitive to time-to-train. For graphics-intensive workloads, the new instances offer up to 1.8x increase in graphics performance and up to 2x video transcoding capability over the previous generation G3 instances. AWS Releases EC2 G4 Instances With NVIDIA T4 Tensor Core GPUs
  6. For one of my four Intel 540s Series 1 TB SATA 6 Gbps SSD that failed (exhausted the write endurance) = 446.4TB over the course of 8392 power-on hours (~350 days). Attached below is the SMART report that was exported from the Intel SSD Toolbox and imported into Excel. (And I have four of these drives. All of them wore through the write endurance limit. My LOWEST was 314.8 TB. What's shown is my "worst case/worst drive".) I had also previously worn out a 400 GB Intel 750 Series PCIe 3.0 NVMe x4 HHHL AIC SSD, but I don't have the SMART results/data handy. So...yeah.
  7. What is the best way to test/benchmark a four-drive SSD RAID0 array in Linux? I have a four-drive RAID0 SSD array running in CentOS 7.6.1810 and I am wondering - what do people do/use or what's the best way to test/benchmark SSD RAID arrays? I've tried testing it using two entirely different methodologies and I am getting completely different set of results, so I am trying to see what people used and/or what's the industry best practices? Thank you.
  8. continuum

    850 EVO vs 970 EVO

    What are your source and destination of the transfer? Is the folder/files transferred exactly the same? As far as real-world use feeling exactly the same that doesn't surprise me, typical (even enthusiast) end-user workloads are so light on the disk that while NVMe drives, technically faster, just don't have enough disk I/O in such situations to show off their performance. What does AS-SSD or CrystalDiskMark or something else say? I'm wondering if you have a host based disk cache enabled or something on the 850 EVO... but again I don't want to jump to too many conclusions here.
  9. velketor

    850 EVO vs 970 EVO

    A friend of mine suggested I upgrade my 500GB 850 EVO to a 500GB 970 EVO because the speeds are much faster. I took his advice and made the switch, but I notice no difference in actual boot time of my PC or the time it takes to open and run video games. However, the speed at which I can transfer files is much slower on the new drive. Transferring a folder full of 8GB worth of content took 8 seconds on the 850 EVO and it takes 25 seconds on the 970 EVO. Why would this be? (see below for Benchmark tests) The 850 EVO is from 2017 and the 970 EVO is from 2018. Surely it should be faster, but it isn't. And here are the results from UserBenchmarks: Game 89%, Desk 155%, Work 97% CPU: Intel Core i7-7700K - 89.5% GPU: Nvidia GTX 1070 - 80.2% SSD: Samsung 970 Evo NVMe PCIe M.2 500GB - 363.6% SSD: Samsung 850 Evo 500GB - 1,210.7% RAM: Corsair Vengeance LPX DDR4 3000 C15 4x8GB - 73% MBD: Gigabyte GA-Z170X-Gaming 7 The 850 Evo is somehow 1200% faster than the 970. I don't understand how that is possible!
  10. I think there are some ways that you could improve the value of your testing to the reader, that are based on the setup and applicability of the results in a CPU scaling determination: Core count: You ran the same number of 16 core VMs on all CPUs. Resulting in massive CPU wait state performance impacts by over-provisioning the vCPUs when run on the 7302P (16C/32T) and to a lesser extent the 7402P (24C/48T). This largely explains the massive latency increase on the 7302P. I understand these are your usual tests, but in this instance, I think it would have been very useful to have run this test with multiple VM counts, such as at 1, 2, 4 and 8 SQL VMs and then we could produce a SQL performance per $ metric as a result. As a result, the review is more an exercise in showing the impact in the VMware world on CPU over-provisioning thresholds when dealing with processor-intensive workloads. As you allude to in the closing paragraphs, many people are interested in these CPUs in single socket designs to save on VMware licensing costs. I would love to see a review of the 7402P, 7502P and 7702P models against systems with two comparable Intel dual-socket CPUs. Such as the 7502P vs two Gold 6242 Xeons. If you can run those tests and 2, 4, 8 and 16 VMs plus add in your VDI testing, that would be an extremely valuable review to many of your readers. Thank you for all your hard work on providing independent news and reviews! Storage review is one of my frequent visits, so keep it up!
  11. Last week
  12. The new Intel Stratix 10 DX FPGAs are the first, and so far only, FPGAs designed to support Intel Ultra Path Interconnect (Intel UPI). Intel is claiming that when paired with future Intel Xeon processors that also support UPI, the new cards will have 37% lower latency compared to cards that can only communicate over PCIe-4. Like all Statix 10 products, the DX cards are built on a 14nm tri-gate architecture. Intel Ships Early Access Stratix 10DX FPGAs
  13. Most of the improvements in this OpenStack Platform 15 look to be security related, although several straddle the line between security and performance. OpenStack 15 does provide support for NVIDIA GPU/vGPUs to support the growing market demand for artificial intelligence (AI) and machine learning (ML) workloads. One of the bigger performance wins comes through improved crypto offload for IPSec VPN to NICs, allowing companies to implement robust security while retaining better network performance. Another improvement which straddles the line between a security and performance improvement is the enhanced TCP stack which should improve application uptime as well as workload security. Another source of potential performance gain is increased backend support for dedicated hardware cryptographic solutions offering more secure messaging while freeing up more flexible system resources. Red Hat Releases OpenStack Platform 15
  14. Kubernetes is an open-source container-orchestration system and has been widely popular since it came out. As more cloud native applications come out, Kubernetes becomes more appealing. And now almost every major vendor has some form of support of the system. Kubernetes 1.16 Released
  15. Synology started nearly 20 years ago with its core mission being to provide data management and security tools for an ever-evolving technology landscape. That company feels confident that it has stuck to this mission over the years while evolving along the way with an industry that is almost always in flux. Synology went on to prove this point by showcasing its next generation files server technology and a demonstration of its license-free protection suite Active Backup. The company spoke about its public cloud service, C2, and its hybrid cloud model for data management and Synology’s cloud data protection service, C2 Backup. Synology 2020 NYC Sees Several Announcements
  16. Gigabyte has a strong partnership with AMD and is a leading systems vendor for the EPYC 7000 Series platforms. With these six new systems being the second wave of lineup released, the GIGABYTE EPYC 7002 portfolio totals nearly 30 systems and includes rack mount servers, GPU servers, and high density multi-node servers. Offering a large and diverse portfolio GIGABYTE looks to be a top vendor in the AMD EPYC 7002 Series system market. GIGABYTE Expands AMD EPYC Rome Server Systems
  17. The nanoEDGE is a small form factor SDDC appliance that runs the full VMware vSphere, VMware vSAN, and VMware NSX in a low cost, small physical volume appliance. The nanoEDGE provides enough compute to run 25-50 VMs. The nanoEDGE is described as 2.5 servers: 2 servers provide vSAN and NSX and the third (0.5) server provides the vSAN Witness Appliance. This allows the solution to utilize 10GbE back to back connectivity for vSAN and vMotion traffic while keeping costs in check. VMware nanoEDGE A Small, Low Cost SDDC
  18. Together the companies in their joint effort provide cost-effective backup and recovery for both on-premises and cloud environments. Veritas NetBackup provides snapshot-based data protection. Integrating NetBackup into Pure Storage’s FlashArray has helped customers guard against ransomware and data loss. Pure Storage integration of FlashBlade into Veritas enables data protection and rapid restore for critical workloads. Additionally, customers can leverage FlashBlade as a Data Hub for streaming analytics and AI cluster projects with Veritas NetBackup providing data protection. Veritas & Pure Storage Collaboration Expansion
  19. As we said at the launch of the new CPUs, the architecture of the second-generation AMD EPYC is all about performance. It is the world’s first high-performance x86 7nm server CPU. AMD has claimed to improve the CPU core execution engine over the original “Zen” core. Additionally, EPYC Rome has twice the density, 1.25x the frequency (at the same power), and uses half the power (at the same performance of the previous model). AMD goes on to state that there is more on-chip bandwidth in all compute modes. The new CPUs also have high memory speed and bandwidth with a maximum theoretical memory bandwidth of 410GB/s (45% higher than Intel Xeon 2nd gen Scalable CPUs). AMD Has A Rome In Rome Day Focusing On EPYC 7002
  20. LOST6200

    Looking for File Eraser Program

    CCleaner has a free version that should provide several versions of drive wipes. You can also try DBan if you want to boot from a disc or make a USB bootable drive.
  21. Oracle’s big thing over the last few years has been autonomous. Chiefly its Autonomous Database (that has seen very good worldwide adoption) but also Oracle Autonomous Transaction Processing and Oracle Autonomous Data Warehouse. For its Autonomous Database, Oracle is expanding the capabilities once more. The company is now offering a dedicated deployment option for Oracle Autonomous Database. This option will give customers high levels of security, reliability, and operational control regardless of the class of database workload. Oracle Announces New Innovations At OpenWorld
  22. Vertica analytics platform in Eon Mode for Pure Storage is a mouthful of words, so let's pull it apart and see what each piece means. Vertica is a set of tools to analyze information stored in an SQL database. Eon Mode is a set of configurations for this set of tools. What makes Eon Mode special is that it allows Vertica customers to separate compute resources from data storage and better control costs. Previously this mode, and thus the cost savings were focused on the AWS ecosystem, but now Micro Focus is extending the same cost savings opportunities to Pure Storage. Micro Focus Releases Vertica in Eon Mode for Pure Storage
  23. The AMD EPYC 7002 came out with the promise of more performance with fewer sockets. Now that VMware fully supports the new CPUs, we continue with our testing to show the promise of the new processors. For this review we were able to test three distinct AMD EPYC CPUs in a traditional VMware environment. This testing wasn’t to show which is better, that is typically the one with the highest core count. Instead we tested the different CPUs for readers to get an idea what to expect, so they can better choose the processor that is right for their use case. AMD EPYC Rome VMware Review (7702P, 7402P, and 7302P)
  24. The partnership allows both companies to reach across the aisle and take advantage of each other’s technology. For instance: Oracle becomes a partner in the VMware Cloud Provider Program and Oracle Cloud VMware Solution will be sold by Oracle and its partners. This Oracle Cloud solution is built on VMware Cloud Foundation and is said to deliver a full stack SDDC that includes VMware vSphere, NSX, and vSAN. Now customers should be able to seamlessly migrate workloads between on-premise environments and Oracle Cloud. VMware Cloud Foundation On Oracle Cloud Infrastructure Announced
  25. Of the five new PowerEdge Servers announced, only the two single-socket servers are currently available. The R6515 is a 1U rack server with a single socket. The R7515 is a similar design, scaled up to a 2U form factor. Dell EMC expects the R6525 and C6525 servers next month. The PowerEdge R6525 is a 1U rack server with two CPU sockets. We have very little solid information on the C6525, but we do know it also has two CPU sockets. It appears to be a 2U form factor. The only other c-series PowerEdge server currently available is the c4140, a 1U form factor rack server so it is difficult to make intelligent guesses as to its exact specs, but it is probably safe to anticipate a design that supports a relatively large number of GPUs or other ancillary processing cards like FPGAs. The fifth and final server also has the vaguest availability date, being projected out to sometime next year. The R7525 will be a two CPU rack server in a 2U form factor. Dell Releases EPYC Rome PowerEdge Servers
  26. EDGEMATRIX will initially focus on the Japanese market, where it has already raised $9 million in Series A funding. Cloudian management is betting heavily on the company’s success with both Cloudian CEO and co-founder Michael Tso and Cloudian board director Jonathan Epstein investing in it. The new company plans to leverage Cloudian’s HyperStore platformto manage large data sets. EDGEMATRIXwill also be inheriting the work Hiroshi Ohta, Cloudian co-founder, and Cloudian teams have done over the past two years to provide AI solutions for such use cases as: targeted advertising using video analysis of the make, model and year of cars on a highway to display targeted ads on a nearby billboard; quality inspection by identifying and removing products that failed to get wrapped in an automated assembly line process; and automated traffic congestion management by counting cars on a road as a basis for changing turn lanes to improve traffic flow. Cloudian Launches Edge AI Subsidiary
  27. Oracle Exadata X8M is a combination of industry-standard Intel 2nd generation Xeon Scalable processors, Intel Optane DC persistent memory, and 100 gigabit RoCE networking. All of this performance-based hardware is designed to support demanding Online Transaction Processing (OLTP), analytics, and mixed workload database requirements, as well as database consolidation and in-database machine learning. Several activities involved in the above are extremely sensitive to latency including high-frequency stock trading, IoT data processing, real-time fraud and intrusion detection, financial trading and applications requiring real-time human interactions. With the integration of Optane, users should see latency improvements in all of the above. Oracle Integrates Intel Optane Into Its Exadata X8M
  28. The release of the ThinkStation P920 Rack offers professional workstation performance in a fully scalable and secure 2U form factor. Leveraging low latency remote access technologies, users are given true freedom by being able to access extreme computing power anywhere and collaborate uninterrupted. This ThinkStation joins their portfolio of P series workstations and is the first rack-optimized form factor in the series. The P920 Rack offers a dual processor platform with Intel Xeon Scalable processors, NVIDIA enterprise grade graphics, up to 3TB of ECC memory, and up to 112TB of enterprise storage. Additionally, enjoy 99.99% availability as it is backed by hallmark ThinkStation brand reliability with hot swappable and redundant features and datacenter-class security. Lenovo ThinkStation P920 Rack Announced
  1. Load more activity