Oblivus United States Chicago Intel Xeon Scalable Virtual Machine Test and Review
We will comprehensively evaluate the Oblivus United States Chicago Intel Xeon Scalable Shared vCPU VPS Server. The server, fuelled by an Intel(R) Xeon(R) Gold 6134 CPU operating at 3.20GHz, offers promising specifications: 2 CPU Cores, 8 GB of RAM, 40 GB NVMe SSD Storage, a high-speed 10Gbps Network Port, Unlimited Traffic, along with 1x IPv4 address.
We will run core metrics tests such as the Basic Benchmark, Yabs, and Unix Benchmark. We will use the Global Node Test for in-depth network assessment, and latency measurement will be undertaken via the Global Ping Test.
Moreover, for evaluating region-specific accessibility, we will be employing the Media Unlock Test. Furthermore, we will use the Cloudflare Internet Test to ratify the connection quality. We will leverage tools like PageSpeed Insights and GTmetrix Performance for browser-based performance.
About Oblivus
Oblivus is a next-generation GPU cloud platform that offers a variety of features like virtual servers, storage, networking, and more, that allows engineers to train and deploy machine learning and deep learning models. It was founded in 2022 and is based in London, United Kingdom.
Oblivus Cloud is available from three distinct regions across the United States, ensuring low-latency access to accelerated computing. These regions are NYC1 (New York City), CHI1 (Chicago), and LAS1 (Las Vegas). Each data center complies with ISO 27001, Type 2 SOC 1, Type 2 SOC 2, and HIPAA standards.
Oblivus offers industry-leading transparent pricing for both GPU and CPU instances. GPU instances start at $0.27/hr for a Quadro RTX 4000 and go up to $2.41/hr for an A100 80GB PCIE. CPU instances are priced based on the requested vCPU cost, with the RAM cost included per vCPU price. Oblivus also offers discounts for workloads with predictable usage patterns through reserved instances, which can result in up to 50% better pricing.
In addition to their current billing system, Oblivus has two automatic systems to provide convenience for their users: Auto Top-up for Individual Accounts and Pay-As-You-Go for Business Accounts. Users can access their support section to create a ticket based on their needs and requirements. Oblivus processes every valid ticket quickly and directs it to the relevant department.
Oblivus is a next-generation GPU cloud platform offering various features at industry-leading transparent pricing. With data centers across the United States and convenient billing options, Oblivus is an excellent choice for beginners and professionals who create and manage websites using hosting or VPS servers.
Official Website: https://www.Oblivus.com
Looking Glass
None
BasicBench: Server Profile, I/O Speed, Upload and Download Speed
CPU Model : Intel(R) Xeon(R) Gold 6134 CPU @ 3.20GHz
CPU Cores : 2 @ 3192.498 MHz
CPU Cache : 16384 KB
AES-NI : Enabled
VM-x/AMD-V : Enabled
Total Disk : 38.7 GB (4.9 GB Used)
Total Mem : 7.8 GB (257.9 MB Used)
System uptime : 0 days, 0 hour 10 min
Load average : 0.09, 0.11, 0.10
OS : Ubuntu 22.04.2 LTS
Arch : x86_64 (64 Bit)
Kernel : 5.19.0-45-generic
TCP CC : bbr
Virtualization : KVM
IPv4/IPv6 : Online / Offline
Organization : AS33425 CoreWeave, Inc
Location : Chicago / US
Region : Illinois
----------------------------------------------------------------------
I/O Speed(1st run) : 174 MB/s
I/O Speed(2nd run) : 345 MB/s
I/O Speed(3rd run) : 397 MB/s
I/O Speed(average) : 305.3 MB/s
----------------------------------------------------------------------
Node Name Upload Speed Download Speed Latency
Speedtest.net 3931.30 Mbps 6209.25 Mbps 1.68 ms
Los Angeles, US 1509.47 Mbps 6371.49 Mbps 61.42 ms
Dallas, US 3551.90 Mbps 4925.84 Mbps 21.32 ms
Montreal, CA 920.49 Mbps 933.89 Mbps 23.72 ms
Paris, FR 946.25 Mbps 4275.70 Mbps 93.23 ms
Amsterdam, NL 915.86 Mbps 2225.29 Mbps 99.57 ms
Shanghai, CN 727.63 Mbps 2861.27 Mbps 189.84 ms
Nanjing, CN 469.21 Mbps 2716.68 Mbps 215.00 ms
Hongkong, CN 6.72 Mbps 2.81 Mbps 229.13 ms
Singapore, SG 372.10 Mbps 1348.41 Mbps 238.74 ms
Tokyo, JP 394.22 Mbps 1821.94 Mbps 150.27 ms
----------------------------------------------------------------------
Finished in : 5 min 38 sec
Timestamp : 2023-08-07 05:43:27 UTC
YabsBench: Disk Speed, iperf3 Network Speed, Geekbench 6 Benchmark
IPv4 Network Information:
---------------------------------
ISP : CoreWeave, Inc
ASN : AS33425 CoreWeave, Inc
Host : RELX inc
Location : Chicago, Illinois (IL)
Country : United States
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 232.96 MB/s (58.2k) | 2.90 GB/s (45.3k)
Write | 233.58 MB/s (58.3k) | 2.91 GB/s (45.5k)
Total | 466.55 MB/s (116.6k) | 5.81 GB/s (90.8k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 8.27 GB/s (16.1k) | 7.07 GB/s (6.9k)
Write | 8.71 GB/s (17.0k) | 7.54 GB/s (7.3k)
Total | 16.98 GB/s (33.1k) | 14.61 GB/s (14.2k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | Ping
Clouvider | London, UK (10G) | 1.98 Gbits/sec | 852 Mbits/sec | 92.1 ms
Scaleway | Paris, FR (10G) | 5.40 Gbits/sec | busy | 96.6 ms
NovoServe | North Holland, NL (40G) | busy | busy | 99.4 ms
Uztelecom | Tashkent, UZ (10G) | 2.24 Gbits/sec | 869 Mbits/sec | 193 ms
Clouvider | NYC, NY, US (10G) | 8.07 Gbits/sec | busy | 23.9 ms
Clouvider | Dallas, TX, US (10G) | 9.05 Gbits/sec | 4.55 Gbits/sec | 21.3 ms
Clouvider | Los Angeles, CA, US (10G) | 3.12 Gbits/sec | 3.85 Gbits/sec | 58.8 ms
Geekbench 6 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 1164
Multi Core | 2010
Full Test | https://browser.geekbench.com/v6/cpu/2176510
YABS completed in 13 min 57 sec
UnixBench: CPU Performance
------------------------------------------------------------------------
Benchmark Run: Mon Aug 07 2023 06:10:40 - 06:38:47
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 40586828.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 7129.8 MWIPS (9.4 s, 7 samples)
Execl Throughput 2562.1 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 491827.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 125789.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1604620.0 KBps (30.0 s, 2 samples)
Pipe Throughput 593495.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 61878.1 lps (10.0 s, 7 samples)
Process Creation 5859.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8080.7 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1486.1 lpm (60.0 s, 2 samples)
System Call Overhead 306879.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 40586828.9 3477.9
Double-Precision Whetstone 55.0 7129.8 1296.3
Execl Throughput 43.0 2562.1 595.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 491827.6 1242.0
File Copy 256 bufsize 500 maxblocks 1655.0 125789.3 760.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 1604620.0 2766.6
Pipe Throughput 12440.0 593495.4 477.1
Pipe-based Context Switching 4000.0 61878.1 154.7
Process Creation 126.0 5859.4 465.0
Shell Scripts (1 concurrent) 42.4 8080.7 1905.8
Shell Scripts (8 concurrent) 6.0 1486.1 2476.8
System Call Overhead 15000.0 306879.2 204.6
========
System Benchmarks Index Score 885.5
------------------------------------------------------------------------
Benchmark Run: Mon Aug 07 2023 06:38:47 - 07:06:58
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 80619127.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 14088.9 MWIPS (9.8 s, 7 samples)
Execl Throughput 4411.7 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 734172.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 196498.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2300875.7 KBps (30.0 s, 2 samples)
Pipe Throughput 1175119.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 256976.9 lps (10.0 s, 7 samples)
Process Creation 12166.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 11376.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1592.2 lpm (60.0 s, 2 samples)
System Call Overhead 575629.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 80619127.4 6908.2
Double-Precision Whetstone 55.0 14088.9 2561.6
Execl Throughput 43.0 4411.7 1026.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 734172.0 1854.0
File Copy 256 bufsize 500 maxblocks 1655.0 196498.5 1187.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 2300875.7 3967.0
Pipe Throughput 12440.0 1175119.6 944.6
Pipe-based Context Switching 4000.0 256976.9 642.4
Process Creation 126.0 12166.2 965.6
Shell Scripts (1 concurrent) 42.4 11376.5 2683.1
Shell Scripts (8 concurrent) 6.0 1592.2 2653.6
System Call Overhead 15000.0 575629.6 383.8
========
System Benchmarks Index Score 1586.3
======= Script description and score comparison completed! =======
Global Node Test
## CDN Speedtest
CacheFly : 478.61 MiB/s | 3828.87 Mbps | ping 2.350ms
Gdrive : 56.21 MiB/s | 449.69 Mbps | ping 2.513ms
## North America Speedtest
Softlayer, Washington, USA : 43.73 MiB/s | 349.84 Mbps | ping 21.607ms
SoftLayer, San Jose, USA : 46.86 MiB/s | 374.88 Mbps | ping 48.131ms
SoftLayer, Dallas, USA : 0 B/s | N/A | ping error!
Vultr, New Jersey, USA : 201.07 MiB/s | 1608.54 Mbps | ping 23.411ms
Vultr, Seattle, USA : 97.12 MiB/s | 776.95 Mbps | ping 51.456ms
Vultr, Dallas, USA : 225.77 MiB/s | 1806.17 Mbps | ping 22.009ms
Vultr, Los Angeles, USA : 59.39 MiB/s | 475.10 Mbps | ping 58.165ms
Ramnode, New York, USA : 89.87 MiB/s | 718.97 Mbps | ping 20.487ms
Ramnode, Atlanta, USA : 90.31 MiB/s | 722.47 Mbps | ping 21.966ms
## Europe Speedtest
Vultr, London, UK : 54.30 MiB/s | 434.43 Mbps | ping 92.278ms
LeaseWeb, Frankfurt, Germany : 36.55 KiB/s | 0.29 Mbps | ping 100.319ms
Hetzner, Germany : 599.97 KiB/s | 4.69 Mbps | ping 110.577ms
Ramnode, Alblasserdam, NL : 47.55 MiB/s | 380.39 Mbps | ping 100.025ms
Vultr, Amsterdam, NL : 49.97 MiB/s | 399.74 Mbps | ping 98.851ms
EDIS, Stockholm, Sweden : 441 B/s | N/A | ping 114.486ms
OVH, Roubaix, France : 406 B/s | N/A | ping 94.073ms
Online, France : 48.69 MiB/s | 389.53 Mbps | ping 96.358ms
Prometeus, Milan, Italy : 335 B/s | N/A | ping 113.578ms
## Exotic Speedtest
Sydney, Australia : 14.83 MiB/s | 118.65 Mbps | ping 234.167ms
Lagoon, New Caledonia : 224 B/s | N/A | ping 216.431ms
Hosteasy, Moldova : 8.43 MiB/s | 67.46 Mbps | ping 151.411ms
Prima, Argentina : 832.62 KiB/s | 6.50 Mbps | ping error!
## Asia Speedtest
SoftLayer, Singapore : 1.42 MiB/s | 11.32 Mbps | ping 212.976ms
Linode, Tokyo, Japan : 10.20 MiB/s | 81.63 Mbps | ping 147.656ms
Linode, Singapore : 7.64 MiB/s | 61.11 Mbps | ping 225.273ms
Vultr, Tokyo, Japan : 32.84 MiB/s | 262.74 Mbps | ping 153.543ms
Benchmark finished in 95 seconds
Global Ping Test
Query: ping 198.176.96.152, interval: 5s, dense mode: no
Probing 198.176.96.152 – ARIN /23 AS33425 COREWEAVE – United States, Illinois, Chicago
Media Unlock Test
** Test Starts At: Mon Aug 7 06:04:16 UTC 2023
** Checking Results Under IPv4
--------------------------------
** Your Network Provider: Coreweave (198.176.*.*)
============[ Multination ]============
Dazn: Yes (Region: US)
HotStar: Yes (Region: US)
Disney+: Yes (Region: US)
Netflix: Yes (Region: US)
YouTube Premium: Yes
Amazon Prime Video: Yes (Region: US)
TVBAnywhere+: Yes
iQyi Oversea Region: US
Viu.com: No
YouTube CDN: Dallas, TX
Netflix Preferred CDN: Chicago, IL
Spotify Registration: No
Steam Currency: USD
ChatGPT: Yes
=======================================
No IPv6 Connectivity Found, Abort IPv6 Testing...
Testing Done!CloudFlare Internet Test
PageSpeed Insights
GTmetrix Performance
Admin Control Panel
Oblivus does offer a flexible infrastructure where switching between GPU and CPU virtual machines is easy. You can deploy, modify, and rescale your virtual machines according to your needs without any hassle.
Server Pricing
Oblivus offers industry-leading transparent pricing for both GPU and CPU instances. GPU instances start at $0.27/hr for a Quadro RTX 4000 and go up to $2.41/hr for an A100 80GB PCIE. CPU instances are priced based on the requested vCPU cost, with the RAM cost included per vCPU price. Oblivus also offers discounts for workloads with predictable usage patterns through reserved instances, which can result in up to 50% better pricing.
In addition to their current billing system, Oblivus has two automatic systems to provide convenience for their users: Auto Top-up for Individual Accounts and Pay-As-You-Go for Business Accounts. Users can access their support section to create a ticket based on their needs and requirements. Oblivus processes every valid ticket quickly and directs it to the relevant department.
Official Website: https://www.Oblivus.com
Coupon
The merchant did not provide
Customer Support
Oblivus is committed to creating a straightforward, transparent system to minimize user problems. In addition to their service, they have created separate documents for API and company service usage.
However, if you have any problems or suggestions, they also offer an integrated support service. Users can access their support section to create a ticket based on their needs and requirements. Oblivus processes every valid ticket quickly and directs it to the relevant department.
Conclusion
Looking at the data compiled from the evaluation, there are a few salient points to note. The I/O speed demonstrated marked improvements over multiple runs, starting from 174 MB/s and climbing to its peak performance at 397 MB/s, reaching an average speed of 305.3 MB/s. Regarding networking, the server performed excellently in various locations, with the highest download speed recorded at 6371.49 Mbps in Los Angeles, US, and the lowest latency at 1.68 ms via Speedtest.net.
Further assessment with the mixed R/W (50/50) disk speed test provided impressive results, with read/write speeds ranging from hundreds of MB/s to a scale of GB/s across different block sizes. With a 4k block size, the read and write speeds were 232.96 MB/s and 233.58 MB/s, respectively, while with a 1m block size, the speeds soared to 7.07 GB/s for read and 7.54 GB/s for write operations. Furthermore, the Geekbench 6 Benchmark Test, widely acclaimed for assessing performance across various workload types, yielded a single-core score of 1164 and a multi-core score of 2010.
As for the UNIX Benchmark test, the server, running one parallel copy of tests, achieved an impressive system benchmark index score of 885.5, while with two parallel copies of tests, the score heightened to 1586.3. The higher the index score, the better the overall performance, and the benchmark revealed noteworthy performance, particularly on ‘File Copy 4096 bufsize 8000 max blocks’ and ‘Dhrystone 2 using register variables’.
In conclusion, the Oblivus United States Chicago Intel Xeon Scalable Shared vCPU VPS Server delivers first-rate performance across various benchmarks. From I/O speed to network latency and disk speed tests, all results suggest a high-performing, robust server capable of handling heavy workloads. Whether you’re considering single or multi-core operations, this server has demonstrated proficiency. It will be a solid choice for today’s digital infrastructure’s intensive and demanding needs.
Official Website: https://www.Oblivus.com
Related Posts
