Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Introducing aurologic.com - Cloud Servers starting at 5.60€ ($5.89) - DDoS-Protection included
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Introducing aurologic.com - Cloud Servers starting at 5.60€ ($5.89) - DDoS-Protection included

jh_aurologicjh_aurologic Member, Patron Provider
edited October 2023 in Offers

After a year of hard work, we‘re pleased to introduce you aurologic.com as new brand, targeting professionals looking for reliable Cloud- and Dedicated Servers hosted on a modern, responsive platform.

Our Stack - Tons of features

Application Stack

Over the course of the past 18 months, we built our matchless cloud- and dedicated server platform. Utilizing a modular microservices architecture, we‘re able to deliver servers within under 10 seconds, while any aspect from collecting racked servers, creating offers and provisioning them, is automated.

Own datacenter

We are operating most services within our owned 800qm² datacenter in the Frankfurt region with up to 240 racks space and diverse darkfiber connectivity to Frankfurt am Main, which is about 20km distant from Interxion/Equinix. Further regions such as Helsinki, Amsterdam and Los Angeles are planned for the future.

Own network

AS30823 is owned and operated by ourself, offering us with the great advantage of flexibility and quality optimizations. Our Arista and Juniper based network architecture consists of dozens of 100Gbps interfaces with connections to premier carriers like Core-Backbone, RETN and CDN77, including private peerings such as 2x 100Gbps Google, 100Gbps Hetzner, 100Gbps myLoc and many more. We constantly optimize our mix, getting a private peering ready whenever possible.

In Frankfurt we run points of presence at Interxion FRA16 and Equinix FR7, from where our n*100Gbps metroring feeds our datacenter redundantly. Using EVPN-VXLan, we're able to provide customers even with fail-safe Layer2 Transport from 1Gbps, 10Gbps, 40Gbps, 100Gbps to multiple of those interfaces.

With direct connectivity to DTAG, Vodafone, O2 and a lot others, through our carriers, our network is a great fit for low latency applications at an affordable, low, bandwidth price.

Looking Glass can be found on https://lg.aurologic.com/.

DDoS-Protection

Using our self developed XDP based DDoS-Protection called flowShield, on Dualport 100Gbps Nvidia Mellanox ConnectX-6 Smart NICs, we‘re able to filter Network and Application aimed attacks before they reach our core network. This is included with all our services free of charge, while customization is possible within our interface using flexrules, flowrules as well as custom thresholds.

High Bandwidth Nodes

Cloud Compute Nodes are based on two 2x 40Gbps (160Gbps total) Nvidia Mellanox network cards, featuring our own XDP development for optimal performance and security through isolation by so called 'firewalld' microservice, attaching to any Cloud Server. Each cloud compute rack features at least four switches, two n+1 redundant ones for public/private network and the same also for storage connectivity.

This way, we‘re able to offer best-in-class congestion free network connectivity at any time of the day.

Cloud-Init Support

Using our cloud-init metadata service, customers can freely define how their instances are spun-up and even use custom golden images for quick mass deployment. Available operating system images are cloud-init ready Cloud Images.

Tripple Replicated Storage

CEPH based storage provides tripple replication of customer data across three physical storage nodes, offering highest data safety and flexibility. In case we need to upgrade, we can just do it by adding more CEPH nodes and disks, without affecting our customers operation.

RESTful API

Similar to our backend stack, our API consists of a modern architecture, allowing customers to write their own automation with our services. You may find our api docs on apidoc.aurologic.com useful.

Modern Interface

Our VueJS based Interface allows full control over services and DDoS-Protection. All actions are asynchronous and immediate, allowing for almost zero waiting time.

Currently available Operating System Images

Installation usually takes about ~10-20 seconds for any of these:

  • Alma Linux 8/9
  • CentOS 7.9
  • Debian 11/12
  • Rocky Linux 8/9
  • Ubuntu 20.04 / 22.04 LTS

Our offers

Cloud Servers starting at 5,60€ ($5.89)

  • Intel Xeon Gold - 1 Core @ 2,4GHz+
  • 2GB Ram
  • 30GB CEPH SSD Storage (rsb.ha3)
  • 1Gbit guaranteed connection
  • 20TB monthly traffic included

Dedicated Servers with 10/40/100Gbps connection

  • From E5v2, AMD Ryzen to AMD EPYC or even custom configured Intel Core i9 - we got multiple and different servers available for all needs - bandwidth hungry, on a budget or high performance
  • 100TB monthly traffic included by default
  • 1Gbps flatrate is just 200€ monthly
  • 10Gbps flatrate is just 1200€ monthly
  • From 1Gbps, 10Gbps, 40Gbps to 100Gbps connection
  • Secured IPMI Access down to Remote Console and Virtual Media
  • Like Cloud Servers - BGP available at 25,00€ monthly / 25€ setup - tunneling is allowed

Planned features on our roadmap

  • Additional disk images as bookable addon
  • Daily encrypted off-site backup (free-of-charge)
  • Resizing of existing images up to 10TB
  • 10Gbit fully usable uplinks with Cloud Servers
  • Upgrade/Downgrade of Instances
  • AMD EPYC and Ryzen offers - coming 11/2023
  • Private Networking (VXLan) between Instances
  • Automated BGP-Sessions

Questions, Thoughts? Just ask me here or hit us a message within our live chat, happy to answer them all! :)

«13

Comments

  • Looking forward to EPYC and Ryzen offers. <3

    Thanked by 1jh_aurologic
  • Carlin0Carlin0 Member
    edited October 2023

    https://lg.aurologic.com/ ← doesn't work...

    EDIT : now yes

    Any cheap VPS ?

  • jh_aurologicjh_aurologic Member, Patron Provider

    @Carlin0 said:
    https://lg.aurologic.com/ ← doesn't work...

    EDIT : now yes

    Any cheap VPS ?

    The most affordable offer starts at 5,60€ while including true redundancy and the said perks described above :)

  • W, any bgp available on the cheapest VPS?

  • jh_aurologicjh_aurologic Member, Patron Provider

    @kait said:
    W, any bgp available on the cheapest VPS?

    BGP is available at 25,00€ monthly / 25,00€ one time with AS30823 on our core routers

  • Intel Xeon Gold - 1 Core @ 2,4GHz+

    Are these dedicated cores?

  • @jh_aurologic said:

    @Carlin0 said:

    Any cheap VPS ?

    The most affordable offer starts at 5,60€ while including true redundancy and the said perks described above :)

    Is not very cheap, best wishes for your business

  • Love the website Design.

  • jh_aurologicjh_aurologic Member, Patron Provider

    @dosai said:
    Intel Xeon Gold - 1 Core @ 2,4GHz+

    Are these dedicated cores?

    These are dedicated, fully usable cores - hence the pricing; Same applies for connectivity, we dont do sharing between resources, any servers have available what you pay for :smile:

  • Nice job, It's nice to see this progress over a long period of time!
    When bring 10Gbps cloud instances?

    Thanked by 1jh_aurologic
  • jh_aurologicjh_aurologic Member, Patron Provider

    @Matix8981 said:
    Nice job, It's nice to see this progress over a long period of time!
    When bring 10Gbps cloud instances?

    We're working on these, especially as we want to offer a really competitive service - the idea is to implement forwarding in XDP to allow real 10Gbps line-speed, other than most competitors do - apart from large cloud providers :)

    At this point, I think 10Gbps as bookable addon will become available within the next eight weeks together with some of the other features mentioned on our list.

    Thanked by 1Matix8981
  • LeviLevi Member
    edited October 2023

    Mmm... CEPH? Someone please yab them. Have a bad notion about that i/o...

  • tentortentor Member, Patron Provider

    @jh_aurologic said: the idea is to implement forwarding in XDP to allow real 10Gbps line-speed

    I doubt that XDP is sufficient to accomplish such crazy goal, I think you'll need to use something like this patch as well.

  • jh_aurologicjh_aurologic Member, Patron Provider
    edited October 2023

    @tentor said:

    @jh_aurologic said: the idea is to implement forwarding in XDP to allow real 10Gbps line-speed

    I doubt that XDP is sufficient to accomplish such crazy goal, I think you'll need to use something like this patch as well.

    It's just about the forwarding (basically fdb lookup + xdp redirect) which is going to be ported to XDP for guest instances. There is a enormous performance benefit over classic linux bridging, which is typically slow due to all the copy operations involved within the kernel. We have been into netmap for years, last year our DDoS-Protection stack got completely ported to XDP to get rid of the userspace limitations we had; Having quite some experience now, which is beneficial for cloud networking B)

    @LTniger said:
    Mmm... CEPH? Someone please yab them. Have a bad notion about that i/o...

    Should be almost as fast as local disk, blockdevice latency might be a bit higher though; In our environment it's at least three dedicated CEPH nodes with 9000 byte MTU and 40G networking in between at the moment. CephFS is slow though, we use RBD natively with Qemu/KVM - you might find https://ceph.io/en/news/blog/2022/qemu-kvm-tuning/ interesting :D

  • tentortentor Member, Patron Provider
    edited October 2023

    @jh_aurologic said:

    @tentor said:

    @jh_aurologic said: the idea is to implement forwarding in XDP to allow real 10Gbps line-speed

    I doubt that XDP is sufficient to accomplish such crazy goal, I think you'll need to use something like this patch as well.

    It's just about the forwarding (basically fdb lookup + xdp redirect) which is going to be ported to XDP for guest instances. There is a enormous performance benefit over classic linux bridging, which is typically slow due to all the copy operations involved within the kernel. We have been into netmap for years, last year our DDoS-Protection stack got completely ported to XDP to get rid of the userspace limitations we had; Having quite some experience now, which is beneficial for cloud networking B)

    Well I get you regarding XDP - if properly used it can make huge performance boost, but isn't the highest overhead for VM is a packet copy itself? XDP is cool for host-only stuff, I cannot see any way how it can be used to pass the packet directly into the guest.

  • jh_aurologicjh_aurologic Member, Patron Provider
    edited October 2023

    @tentor said:
    Well I get you regarding XDP - if properly used it can make huge performance boost, but isn't the highest overhead for VM is a packet copy itself? XDP is cool for host-only stuff, I cannot see any way how it can be used to pass the packet directly into the guest.

    virtio tap with 16 rx/tx (combined) queues is kinda fast, you can easily push 6-8mpps from a potent guest - done that in our early days with XDP in our lab. Thats something we would like to cover to avoid bottleneck situations in our production environment - the reason we run each node with two dualport 40G NICs (so to say 80Gbps network + 80Gbps storage) is not just out of coincidence, it's to ensure stability on network level, same applies for the XDP based forwarding with 10Gbps instances.

    Here in our datacenter, we have now Arista 100G Chassis with 128x 100G ports, newish (market related) 100G Bidi ER4 (40km) transceivers and probably as of Q1/2024 the first 100G DWDM Duplex ones. The same level of capacity is something I'd like to keep in our Cloud Environment, thats what customers pay us for - decent capacity, while aiming for a rock solid service.

    Thanked by 1tentor
  • very cool, and props for accepting crypto

  • DPDP Administrator, The Domain Guy

    What's the IPv6 allocation for the Cloud Servers?

  • # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-09-06                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri Oct  6 20:43:05 UTC 2023
    
    Basic System Information:
    ---------------------------------
    Uptime     : 0 days, 0 hours, 1 minutes
    Processor  : Intel Xeon Processor (Skylake, IBRS)
    CPU cores  : 1 @ 2394.374 MHz
    AES-NI     : � Enabled
    VM-x/AMD-V : � Enabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 27.0 GiB
    Distro     : Ubuntu 22.04.3 LTS
    Kernel     : 5.15.0-1042-kvm
    VM Type    : KVM
    IPv4/IPv6  : � Online / � Offline
    
    IPv4 Network Information:
    ---------------------------------
    ISP        : aurologic GmbH
    ASN        : AS30823 aurologic GmbH
    Host       : UAB REVONIA
    Location   : London, England (ENG)
    Country    : United Kingdom
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ----
    Read       | 41.67 MB/s   (10.4k) | 161.19 MB/s   (2.5k)
    Write      | 41.76 MB/s   (10.4k) | 162.04 MB/s   (2.5k)
    Total      | 83.43 MB/s   (20.8k) | 323.23 MB/s   (5.0k)
               |                      |
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ----
    Read       | 212.73 MB/s    (415) | 209.10 MB/s    (204)
    Write      | 224.03 MB/s    (437) | 223.03 MB/s    (217)
    Total      | 436.76 MB/s    (852) | 432.13 MB/s    (421)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
    -----           | -----                     | ----            | ----            | ----
    Clouvider       | London, UK (10G)          | 481 Mbits/sec   | 536 Mbits/sec   | 13.9 ms
    Scaleway        | Paris, FR (10G)           | 1.08 Gbits/sec  | 851 Mbits/sec   | 12.9 ms
    NovoServe       | North Holland, NL (40G)   | 1.09 Gbits/sec  | 949 Mbits/sec   | 8.02 ms
    Uztelecom       | Tashkent, UZ (10G)        | 1.04 Gbits/sec  | 635 Mbits/sec   | 73.0 ms
    Clouvider       | NYC, NY, US (10G)         | busy            | 174 Mbits/sec   | 80.0 ms
    Clouvider       | Dallas, TX, US (10G)      | 221 Mbits/sec   | 39.8 Mbits/sec  | 226 ms
    Clouvider       | Los Angeles, CA, US (10G) | 960 Mbits/sec   | 59.0 Mbits/sec  | 137 ms
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value
                    |
    Single Core     | 1253
    Multi Core      | 1244
    Full Test       | https://browser.geekbench.com/v6/cpu/2955086
    
    YABS completed in 14 min 7 sec
    
    Thanked by 3maverick dosai Carlin0
  • jh_aurologicjh_aurologic Member, Patron Provider

    @DP said:
    What's the IPv6 allocation for the Cloud Servers?

    With both Cloud- and Dedicated Servers, we just allocate a single IPv6-Address, so to say /128 - any requirement for more? :)

  • Oh. You renamed.
    Any effects on „older“ KVM resources?
    My system seems to be moved over to your new name.

    Also installation of my own ISO (or FreeBSD) is still an option?

    Thanks

  • jh_aurologicjh_aurologic Member, Patron Provider

    @wedge1001 said:
    Oh. You renamed.
    Any effects on „older“ KVM resources?
    My system seems to be moved over to your new name.

    Also installation of my own ISO (or FreeBSD) is still an option?

    Thanks

    We will migrate them off the current GlusterFS/Proxmox infrastructure onto the new environment soon. The current HP Gen8/Gen9 blades these are running on, are kinda power hungry, while Proxmox and GlusterFS isnt something we would like to maintain anylonger. Thats basically monolithic php code and hardware from 2018-2019, reflecting in quite some workload 😬

    ISO imports are already integrated, while not released yet. Its on our list of things to get done, same applies to BSD Images, which need a bit of a testing 😄

    Thanked by 1wedge1001
  • Any lower pricings? Why dedis cost much more than somewhere else?

  • @Carlin0 said:
    Any cheap VPS ?

    any lowend offers?!

  • @jh_aurologic is my fastpipe.io login data still working?

  • @jh_aurologic said: including true redundancy

    What other redundancy is included other than CEPH storage ? If hypervisor completely fails is VM migrated and started automatically on another node ?

  • @hyperblast said:
    @jh_aurologic is my fastpipe.io login data still working?

    confirmed. old fastpipe.io username and password works with aurologic.

  • @hyperblast said:

    @hyperblast said:
    @jh_aurologic is my fastpipe.io login data still working?

    confirmed. old fastpipe.io username and password works with aurologic.

    care to share your login details on here so that we can confirm too?

    Thanked by 1SirFoxy
  • @emgh said:

    @hyperblast said:

    @hyperblast said:
    @jh_aurologic is my fastpipe.io login data still working?

    confirmed. old fastpipe.io username and password works with aurologic.

    care to share your login details on here so that we can confirm too?

    of course, not!

    Thanked by 1emgh
  • but i could share my greencloud access data... just checked... are unfortunately not available. :)

    Thanked by 1emgh
Sign In or Register to comment.