Oblivus United States New York AMD EPYC Rome Virtual Machine Test and Review
We will review the Oblivus cloud server in New York, which uses a powerful AMD EPYC 7402P with 24 cores. This server setup includes a 2-core CPU, 8GB of RAM, 40GB of SSD storage, super-fast 10Gbps internet, unlimited data transfer, and one IPv4 address.
We’ll check its performance using the following:
1. Tests like Basic Benchmark and Yabs to see the server’s raw performance.
2. Unix Benchmark to measure standard operations.
3. Global Node and Ping tests to evaluate network speed and how quickly it responds.
4. Use the Media Unlock Test to check if it can access content from various regions.
5. CloudFlare’s test to judge the quality of the server connection.
6. PageSpeed Insights and GTmetrix will rate how well it handles web content.
Stay tuned for straightforward and detailed insights on this server’s capabilities!
About Oblivus
Oblivus.com is a next-generation cloud service provider that specializes in offering GPU and CPU server products. It’s a platform that revolutionizes the way we think about cloud computing, particularly when it comes to GPU acceleration.
The company aims to provide everyone with cheap, fast, and reliable infrastructure. They aim to make high-performance computing accessible and affordable. With Oblivus, you can expect to save up to 80% on your cloud bill compared to other service providers.
Oblivus boasts an impressive infrastructure equipped to meet various computing requirements. Whether you need one or thousands of GPUs or one vCPU to tens of thousands of vCPUs, Oblivus has got you covered. Their resources are readily available to cater to your needs whenever you need them.
One of the standout features of Oblivus is its flexible infrastructure. Switching between GPU and CPU instances is a breeze with their platform. You can deploy, modify, and rescale your instances according to your needs without any hassle.
Oblivus offers a vast range of NVIDIA GPUs and the latest CPUs in the industry. They provide access to 13 enterprise-grade NVIDIA GPU SKUs and the latest CPUs, ensuring they have the right solution to meet your computing needs. Their GPU instances offer outstanding machine learning performance without breaking the bank.
Official Website: https://www.Oblivus.com
How does Oblivus compare to other cloud providers?
Oblivus Cloud is a unique platform in the cloud service provider landscape. Here’s how it compares to other providers:
- Customization: Unlike many cloud service providers that offer instances with pre-set configurations, Oblivus allows users to modify each major component of their instance. You can scale up to thousands of GPUs and vCPUs within seconds. Oblivus also allows you to switch between GPU and CPU instances as needed.
- Pricing: Oblivus provides affordable, transparent pricing, and up to 80% cheaper than major cloud service providers. This makes it a cost-effective choice for users who require powerful computing resources.
- Ease of Use: Oblivus aims to make high-performance computing straightforward and accessible. They have designed their platform to be user-friendly, even for individuals who may not have extensive knowledge about cloud computing.
- Features: Oblivus Cloud comes packed with a wide range of features. These include fully customizable infrastructure, no quotas or complex verification processes, 10Gbps to 40Gbps public network connectivity, transparent and affordable per-minute-based Pay-As-You-Go pricing with no hidden fees, and more.
- Hardware: Oblivus offers a variety of cutting-edge CPUs and 10 state-of-the-art GPU SKUs. This ensures they have the right solution to meet your computing needs.
Oblivus stands out for its customization options, affordable pricing, ease of use, feature-rich platform, and various hardware options. It’s a compelling alternative to other cloud service providers, particularly for users who require powerful computing resources.
BasicBench: Server Profile, I/O Speed, Upload and Download Speed
CPU Model : AMD EPYC 7402P 24-Core Processor
CPU Cores : 2 @ 2794.748 MHz
CPU Cache : 512 KB
AES-NI : ✓ Enabled
VM-x/AMD-V : ✓ Enabled
Total Disk : 38.7 GB (4.8 GB Used)
Total Mem : 7.8 GB (428.4 MB Used)
System uptime : 0 days, 0 hour 11 min
Load average : 0.20, 0.09, 0.03
OS : Ubuntu 20.04.6 LTS
Arch : x86_64 (64 Bit)
Kernel : 5.15.0-84-generic
TCP CC : bbr
Virtualization : KVM
IPv4/IPv6 : ✓ Online / ✗ Offline
Organization : AS33425 CoreWeave, Inc
Location : New York City / US
Region : New York
----------------------------------------------------------------------
I/O Speed(1st run) : 221 MB/s
I/O Speed(2nd run) : 219 MB/s
I/O Speed(3rd run) : 81.7 MB/s
I/O Speed(average) : 173.9 MB/s
----------------------------------------------------------------------
Node Name Upload Speed Download Speed Latency
Speedtest.net 9373.32 Mbps 9112.25 Mbps 0.49 ms
Los Angeles, US 1373.11 Mbps 4203.21 Mbps 63.94 ms
Dallas, US 2746.90 Mbps 7290.59 Mbps 33.18 ms
Montreal, CA 928.40 Mbps 820.27 Mbps 8.15 ms
Paris, FR 1100.76 Mbps 5541.18 Mbps 79.88 ms
Amsterdam, NL 1161.70 Mbps 4422.59 Mbps 76.80 ms
Shanghai, CN 781.41 Mbps 4739.76 Mbps 205.69 ms
Chongqing, CN 16.80 Mbps 0.61 Mbps 250.37 ms
Mumbai, IN 1062.22 Mbps 3946.32 Mbps 199.21 ms
Singapore, SG 1651.31 Mbps 6346.87 Mbps 247.54 ms
Tokyo, JP 510.62 Mbps 2480.59 Mbps 168.96 ms
----------------------------------------------------------------------
Finished in : 5 min 36 sec
Timestamp : 2023-11-24 00:30:16 UTC
YabsBench: Disk Speed, iperf3 Network Speed, Geekbench 6 Benchmark
Basic System Information:
---------------------------------
Uptime : 0 days, 0 hours, 21 minutes
Processor : AMD EPYC 7402P 24-Core Processor
CPU cores : 2 @ 2794.748 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 7.8 GiB
Swap : 0.0 KiB
Disk : 38.7 GiB
Distro : Ubuntu 20.04.6 LTS
Kernel : 5.15.0-84-generic
VM Type : VM-OTHER
IPv4/IPv6 : ✔ Online / ❌ Offline
IPv4 Network Information:
---------------------------------
ISP : CoreWeave, Inc
ASN : AS33425 CoreWeave, Inc
Location : New York, New York (NY)
Country : United States
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 319.16 MB/s (79.7k) | 10.96 MB/s (171)
Write | 320.01 MB/s (80.0k) | 11.51 MB/s (179)
Total | 639.18 MB/s (159.7k) | 22.47 MB/s (350)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 921.00 KB/s (1) | 6.95 MB/s (6)
Write | 933.00 KB/s (1) | 7.64 MB/s (7)
Total | 1.85 MB/s (2) | 14.59 MB/s (13)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | Ping
Clouvider | London, UK (10G) | 2.68 Gbits/sec | 924 Mbits/sec | 69.8 ms
Scaleway | Paris, FR (10G) | busy | busy | 83.3 ms
NovoServe | North Holland, NL (40G) | 2.45 Gbits/sec | 2.79 Gbits/sec | 76.5 ms
Uztelecom | Tashkent, UZ (10G) | 2.62 Gbits/sec | busy | 171 ms
Clouvider | NYC, NY, US (10G) | 9.18 Gbits/sec | 7.87 Gbits/sec | 1.43 ms
Clouvider | Dallas, TX, US (10G) | 664 Mbits/sec | 1.01 Gbits/sec | 151 ms
Clouvider | Los Angeles, CA, US (10G) | 2.92 Gbits/sec | 2.98 Gbits/sec | 63.2 ms
Geekbench 6 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 1362
Multi Core | 2458
Full Test | https://browser.geekbench.com/v6/cpu/3683765
YABS completed in 14 min 52 sec
UnixBench: CPU Performance, A score higher is better
Benchmark Run: 11 24 2023 01:13:16 - 01:41:35
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 35474687.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4887.7 MWIPS (9.6 s, 7 samples)
Execl Throughput 2502.5 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 579013.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 141225.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1657151.0 KBps (30.0 s, 2 samples)
Pipe Throughput 1195724.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 35077.4 lps (10.0 s, 7 samples)
Process Creation 4235.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2745.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 575.1 lpm (60.1 s, 2 samples)
System Call Overhead 1561487.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 35474687.6 3039.8
Double-Precision Whetstone 55.0 4887.7 888.7
Execl Throughput 43.0 2502.5 582.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 579013.6 1462.2
File Copy 256 bufsize 500 maxblocks 1655.0 141225.8 853.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 1657151.0 2857.2
Pipe Throughput 12440.0 1195724.7 961.2
Pipe-based Context Switching 4000.0 35077.4 87.7
Process Creation 126.0 4235.6 336.2
Shell Scripts (1 concurrent) 42.4 2745.9 647.6
Shell Scripts (8 concurrent) 6.0 575.1 958.4
System Call Overhead 15000.0 1561487.4 1041.0
========
System Benchmarks Index Score 827.2
------------------------------------------------------------------------
Benchmark Run: 11 24 2023 01:41:35 - 02:10:08
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 70135311.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 9522.1 MWIPS (9.8 s, 7 samples)
Execl Throughput 5105.1 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 490876.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 174470.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1396627.4 KBps (30.0 s, 2 samples)
Pipe Throughput 1746300.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 315146.0 lps (10.0 s, 7 samples)
Process Creation 8801.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5977.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1068.7 lpm (60.0 s, 2 samples)
System Call Overhead 2225592.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 70135311.3 6009.9
Double-Precision Whetstone 55.0 9522.1 1731.3
Execl Throughput 43.0 5105.1 1187.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 490876.0 1239.6
File Copy 256 bufsize 500 maxblocks 1655.0 174470.3 1054.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 1396627.4 2408.0
Pipe Throughput 12440.0 1746300.1 1403.8
Pipe-based Context Switching 4000.0 315146.0 787.9
Process Creation 126.0 8801.5 698.5
Shell Scripts (1 concurrent) 42.4 5977.6 1409.8
Shell Scripts (8 concurrent) 6.0 1068.7 1781.2
System Call Overhead 15000.0 2225592.5 1483.7
========
System Benchmarks Index Score 1481.6
======= Script description and score comparison completed! =======
Looking Glass
None
Global Node Test
## CDN Speedtest
CacheFly : 427.35 MiB/s | 3418.80 Mbps | ping 0.544ms
Gdrive : 0 B/s | N/A | ping 19.989ms
## North America Speedtest
Softlayer, Washington, USA : 167.81 MiB/s | 1342.50 Mbps | ping 6.422ms
SoftLayer, San Jose, USA : 43.00 MiB/s | 343.99 Mbps | ping 69.924ms
SoftLayer, Dallas, USA : 0 B/s | N/A | ping error!
Vultr, New Jersey, USA : 46.79 MiB/s | 374.36 Mbps | ping 2.308ms
Vultr, Seattle, USA : 73.86 MiB/s | 590.84 Mbps | ping 66.420ms
Vultr, Dallas, USA : 145.56 MiB/s | 1164.48 Mbps | ping 33.519ms
Vultr, Los Angeles, USA : 74.07 MiB/s | 592.59 Mbps | ping 63.467ms
Ramnode, New York, USA : 106.32 MiB/s | 850.55 Mbps | ping 0.489ms
Ramnode, Atlanta, USA : 90.91 MiB/s | 727.30 Mbps | ping 16.859ms
## Europe Speedtest
Vultr, London, UK : 69.35 MiB/s | 554.79 Mbps | ping 69.634ms
LeaseWeb, Frankfurt, Germany : 38.83 KiB/s | 0.30 Mbps | ping 86.202ms
Hetzner, Germany : 292 B/s | N/A | ping 94.730ms
Ramnode, Alblasserdam, NL : 51.91 MiB/s | 415.32 Mbps | ping 78.038ms
Vultr, Amsterdam, NL : 62.81 MiB/s | 502.51 Mbps | ping 76.959ms
EDIS, Stockholm, Sweden : 485 B/s | N/A | ping 94.222ms
OVH, Roubaix, France : 389 B/s | N/A | ping 81.518ms
Online, France : 55.25 MiB/s | 442.03 Mbps | ping 83.270ms
Prometeus, Milan, Italy : 341 B/s | N/A | ping 103.001ms
## Exotic Speedtest
Sydney, Australia : 20.78 MiB/s | 166.26 Mbps | ping 211.474ms
Lagoon, New Caledonia : 264 B/s | N/A | ping 233.646ms
Hosteasy, Moldova : 7.43 MiB/s | 59.42 Mbps | ping 127.954ms
Prima, Argentina : 4.73 MiB/s | 37.85 Mbps | ping error!
## Asia Speedtest
SoftLayer, Singapore : 3.04 MiB/s | 24.33 Mbps | ping 246.022ms
Linode, Tokyo, Japan : 8.21 MiB/s | 65.71 Mbps | ping 192.537ms
Linode, Singapore : 5.86 MiB/s | 46.85 Mbps | ping 231.198ms
Vultr, Tokyo, Japan : 30.52 MiB/s | 244.13 Mbps | ping 164.919ms
Benchmark finished in 99 seconds
Global Latency Test
Query: ping 216.153.57.140, interval: 5s, dense mode: no
Probing 216.153.57.140 – ARIN /21 AS33425 COREWEAVE – United States, New York
Media Unlock Test
** Test Starts At: Fri Nov 24 00:56:41 UTC 2023
** Checking Results Under IPv4
--------------------------------
** Your Network Provider: Coreweave (216.153.*.*)
============[ 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: New York, NY
Netflix Preferred CDN: Washington DC
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 Cloud’s admin console is designed to be user-friendly and intuitive, making it easy for users to manage their resources. It provides a centralized location where users can monitor and control their instances, whether they’re using GPU or CPU servers.
The console allows users to deploy, modify, and rescale instances according to their needs. This flexibility is a key feature of Oblivus, enabling users to adapt their infrastructure to changing requirements without hassle.
In addition to managing instances, the console provides access to Oblivus’s lightning-fast API. This allows users to program their infrastructure according to their needs.
For companies, Oblivus offers the ability to create an organization. This feature allows multiple users to use the same billing profile while maintaining separate credentials for each user. It simplifies management and enhances security.
Furthermore, Oblivus’s custom operating system, OblivusAI, is included in the console. Based on Ubuntu, it comes with all the necessary machine and deep learning libraries, enabling users to start training their models immediately.
Server Pricing
In terms of pricing, Oblivus offers industry-leading transparent pricing. Their GPU instances start as low as $0.27/hr, and their CPU instances start as low as $0.015/hr. This makes Oblivus an attractive option for cost-effective cloud computing solutions.
Official Website: https://www.Oblivus.com
Coupon
The merchant did not provide
Payments
Oblivus offers several payment methods to cater to a wide range of users. They accept card payments through Revolut, bank transfers, and cryptocurrency payments through Binance Pay. This flexibility in payment methods makes accessing their services convenient for users worldwide.
In terms of billing, Oblivus operates on a Pay-As-You-Go system. This means you only pay for the resources you use, making it a cost-effective solution for many users. They also offer an Auto Top-Up system, a convenient way to ensure your account balance never runs out.
Whether you’re an individual or a business, Oblivus makes it easy to manage your payments and keep track of your usage. Their transparent pricing and flexible payment options stand out in the cloud service provider landscape.
Customer Support
Oblivus is committed to providing excellent customer support. They understand the importance of timely and effective assistance when dealing with cloud services.
Email Support: Customers can typically reach out via email for non-urgent issues or inquiries. This method is useful for complex queries that require detailed explanations.Sales Department: [email protected], Support Department: [email protected]
Ticketing System: A ticketing system can track and manage support requests. Customers can submit a ticket detailing their issue, and the support team will address it in the order received.
Conclusion
In our tests, we first looked at input/output operations, measuring how fast the server reads and writes data. On average, speeds were impressive, clocking in at 173.9 MB/s, though the third run dipped to 81.7 MB/s, suggesting some variability under load.
Network performance was robust in most locations based on our speed tests. However, connectivity to some areas – notably Chongqing, China – showed significantly reduced upload and download speeds, indicating regional network constraints.
The Yabs benchmark, which measures disk read and write operations under varying conditions, showed read and write speeds consistently above 10 MB/s for most block sizes. Still, there was a drop-off with larger blocks, which is not uncommon in mixed-use scenarios.
System performance, measured by UnixBench, showed high double-digit or triple-digit index scores across various metrics, signaling exceptional performance, particularly when both CPUs were fully leveraged.
The server proved its strength in raw numbers. It demonstrated excellent capability in accessing geo-restricted media content across major providers like Netflix, Disney+, and Amazon Prime Video, to name a few. This benefits those intending to use the server for media streaming.
In conclusion, the Oblivus server showcased solid performance in both processing and network capabilities. It proved a reliable option for various uses, from web hosting to media streaming, backed by its strong benchmarking results and media service accessibility tests. This level of performance and flexibility positions it well for businesses or individuals needing a high-performance server with excellent data and network service.
Official Website: https://www.Oblivus.com
Related Posts
