Archive for January 24, 2024

OVHcloud Enhances Its AI Offer with a Complete Range of Innovative Serverless Solutions

Posted in Commentary with tags on January 24, 2024 by itnerd

 OVHcloud, the European Cloud leader, today confirms it ambitions for AI, answering skyrocketing demand from its customer base, through an approach that makes AI simple and affordable. In an unprecedented context where generative AI has arguably taken the world by storm, OVHcloud has set itself on a mission to help customers grow their businesses by providing easy and affordable AI solutions, fueled by the latest GPU, all with the ultimate goal of democratizing AI. To make this AI revolution comes true, OVHcloud’s AI strategy focuses on offering four key items starting with efficient yet powerful compute resources, streamlined datasets, tooling in the form of software and cutting-edge skillsets.


Offering compute resources with a broad range of datacenter GPUs

To help organisations gain access to easy and affordable AI, OVHcloud leverages 20 years of expertise in infrastructure, through its unique vertically integrated model. As such, the Group provides eco-friendly compute resources benefiting from its bespoke watercooling system. Adding to previously announced NVIDIA DGX H100 available to select customers, the Group today launches new Compute GPU instances with 1x to 4x NVIDIA H100 PCIe GPU, high RAM allocation and local high-performance storage. Immediately available, the NVIDIA H100 PCIe joins existing NVIDIA A100 offer as well as new and forthcoming NVIDIA L40S, NVIDIA L4 and NVIDIA A10 options.

For customers in need of dedicated Bare Metal servers with GPU horsepower for use cases like Machine Learning, deep learning or deployment of LLM models, OVHcloud today unveils new HGR-AI-2 references powered by NVIDIA L40S GPU. Available this spring, new Bare Metal HGR-AI-2 leverages NVIDIA L40S GPU that benefits from fourth-generation Tensor Cores and FP8 Transformer Engine providing robust performance for AI workloads both in training and inferencing.

Customers looking for GPU-equipped Bare Metal servers can benefit from the new Bare Metal SCALE GPU launching later this year. With three new references in the SCALE GPU line, the Group is addressing use cases such as 3D Rendering, video processing and data inference all powered by NVIDIA L4 Datacenter GPU. Based on the NVIDIA Ada Lovelace GPU architecture, the NVIDIA L4 is a universal GPU for every workload with enhanced AI and video capabilities providing efficient compute resources for graphics, simulation, data science and data analytics.


Streamlining datasets with OVHcloud Data Platform maximizing data privacy

In addition to these technical resources, OVHcloud is making further data, tooling and skills assets available, in alignment with its vision for AI. The OVHcloud Data Platform is a new end-to-end data platform announced at OVHcloud Summit and provides a low code solution for management of the data pipeline and analytics. It provides customers with a complete and easy experience for their data journey offering the availability to collect (with more than 50 available connectors), store, manage, analyze, and visualize data.

Furthermore, the new solution also ensures that data is secure from unauthorized accesses and breaches while complying with laws and regulations. With the OVHcloud Data Platform, data scientists and business analysts can start new projects in just a few clicks by selecting information of interest in the Data Lakehouse. The solution is aimed at use cases like CRM enhancement or streamlined inventory management.

Providing tooling with serverless solutions for AI and a supporting ecosystem

To further help organizations in their data journey from ideation to production, OVHcloud has developed a complete set of serverless tools with AI Notebooks, AI Training and AI Deploy. OVHcloud AI solutions act as a complete set of tools that are easy to use, and designed to experiment with data, train models and put them into production.

Relying on high specification OVHcloud AI infrastructure, including recently announced NVIDIA H100 PCIe, A100, V100S and upcoming L40S and L4, these high-level software solutions leveraging industry standard technologies. They are designed to quickly and accessibly empower data scientists and machine learning engineers.

Additionally, OVHcloud is developing AI Endpoints, a new serverless solution targeted at developers and integrators. With OVHcloud AI Endpoints, the Group will enable access to diverse and various AI models with simple API endpoints that require little to no knowledge on the user side. Leveraging the Group’s infrastructure, OVHcloud AI Endpoints will fully respect data privacy (in/out) and will provide developers with a playground to test the Endpoints before placing API calls. With access to a number of market leading AI models such as BERT, YOLO or Mistral 7B, OVHcloud AI Endpoints is expected to offer a wide range of models when it launches.

OVHcloud is also launching AI App Builder, which is a fully managed serverless solution providing an easy way to build contextualized generative AI assistants. The tool allows customers to input their private datasets in a secured cloud environment, select cutting-edge LLM with just a single click and finally deploy code ready to use for their application. It is available today as an Alpha version here.

Finally, to support the broader ecosystem in the development of high-level AI skills, OVHcloud is working with a number of integrators and partners. The group is offering webinars and will attend the World AI Cannes Festival taking place in Cannes, France from February 8th to 10th

Pricing and availability

OVHcloud GPU instances based on NVIDIA H100 PCIe are available now on Public Cloud starting at CAD 4.05 per hour. Managed AI solutions based on H100 PCIe are available now at CAD 4.48 per hour (billing granularity per minute).

NVIDIA L40S, L4 and A10 GPU instances will be available this spring on Public Cloud starting at CAD 2.02 per hour for L40S instances and CAD 1.08 per hour for L4. New HGR-AI-2 Bare Metal Servers are expected to be available as build to order references in March in Canada (in Beauharnois and in a new data center located in Cambridge, Ontario), Europe and the United States before reaching Asia. New SCALE-GPU are launching later this year in the same geographical regions.

Resources

Cisco Launches Smart Agent For Cisco AppDynamics

Posted in Commentary with tags on January 24, 2024 by itnerd

Cisco today launched Smart Agent for Cisco AppDynamics, enabling agent lifecycle management, dramatically simplifying application instrumentation for full-stack observability through intelligent agent automation and management, and helping customers onboard new applications faster. Customers can identify out-of-date agents and upgrade them in minutes with an easy-to-use centralized agent management user interface. 

As applications become more distributed, modular and scalable, technologists find themselves managing an explosion in the number of agents. While in many companies, the operations teams are dealing with agents in the tens of thousands, those at large enterprises often handle hundreds of thousands of agents, depending on the number and type of applications being observed. Due to the sheer volume of agents — and the time-consuming and complex nature of updating them — many organizations fail to upgrade to the latest versions in a timely fashion. This leaves them with old and unsupported agents, without the latest functionality, and at risk from security vulnerabilities that have been mitigated in the most recent agent releases.  

IT infrastructure and operations teams require powerful insights to help cut through this increasing complexity, and with Smart Agent for Cisco AppDynamics, customers can: 

  • Simplify the process of installing and upgrading agents: New UI-driven process intelligently automates the installation and upgrade of agents, simplifying the process to just a few clicks. This allows customers to upgrade multiple agents at one time, freeing up resources and talent to focus on the management of application performance. 
  • Access new capabilities faster with push-button upgrades: Smart Agent makes it easy to maintain compliance and upgrade agents on a regular basis, ensuring customers have the latest versions and access to new functionality, while minimizing risk of security vulnerabilities. Smart Agent can roll-back or upgrade agents at scale with speed and agility, allowing operations teams to perform these tasks through the UI rather than manual actions on each host. 
  • Gain valuable application performance and business insights at speed: Complicated process discovery and time-consuming agent deployments can make it challenging to obtain the visibility and insight into application performance that is needed to deliver optimal user experience and business outcomes. Smart Agent automates the new application instrumentation process with a single agent install that discovers all the processes running on the host, and automatically instruments the full-stack with the agents needed, as governed by a company’s own policies. Smart Agent auto-discovery and auto-deployment enables customers to start collecting data and visualize application topography in minutes instead of days. 

Smart Agent for Cisco AppDynamics is now generally available. Customers can visit our Smart Agent webpage for more information or schedule a time to speak with us. The auto-deployment functionality will be generally available in CYQ2 2024.  

Register here for the Smart Agent for Cisco AppDynamics webinar on February 14 to learn more.

Horizon3.ai Publishes POC Exploit For Fortra GoAnywhere MFT Authentication Bypass

Posted in Commentary with tags on January 24, 2024 by itnerd

Horizon3.ai Chief Attack Engineer Zach Hanley and the Horizon3.ai Red Team have just published “CVE-2024-0204: Fortra GoAnywhere MFT Authentication Bypass Deep-Dive,” which includes a proof-of concept-exploit (POC) on the widely-used managed file transfer software along with indicators of compromise (IOCs).

Fortra’s GoAnywhere MFT file transfer software is widely used in finance, finance, healthcare, engineering, gaming, logistics, manufacturing, public sector/government, higher education and other sectors to automate and encrypt data between an organization and its trading partners, centralizing file transfer activity and monitoring while improving costs. 

On January 22, 2024, Fortra published a security advisory on CVE-2024-0204, warning of an authentication bypass in Fortra’s GoAnywhere MFT prior to 7.4.1 that allows an unauthorized user to remotely create an admin user via the administration portal. Customers were made aware of the issue by an internal security advisory post and patch made available on December 4, 2023, in which researchers malcolm0x and Islam Elrfai were originally credited with the discovery. In 2023, file transfer applications were a top target by threat actors.

Links

Horizon3.ai’s “CVE-2024-0204: Fortra GoAnywhere MFT Authentication Bypass Deep-Dive” also includes indicators of compromise (IOCs) and remediation recommendations.

Horizon3.ai “CVE-2024-0204: Fortra GoAnywhere MFT Authentication Bypass Deep-Dive” (January 23, 2024): https://www.horizon3.ai/cve-2024-0204-fortra-goanywhere-mft-authentication-bypass-deep-dive/

Horizon3.ai Proof of Concept for CVE-2024-0204: https://github.com/horizon3ai/CVE-2024-0204

Fortra “FI-2024-001 – Authentication Bypass in GoAnywhere MFT” (January 22, 2024): https://www.fortra.com/security/advisory/fi-2024-001

Aptum Forms Partnership with AVANT

Posted in Commentary with tags on January 24, 2024 by itnerd

 Aptum, a global cloud solutions provider specializing in technology consulting and managed services, today announced a strategic partnership agreement with AVANT, a platform for IT decision-making and a premier distributor for next-generation technologies. 

This collaboration makes Aptum’s comprehensive cloud solutions available to AVANT clients, and enables the companies to align their sales and support organizations in the U.S., Canada, and the UK.   

Now with more than a decade of experience, AVANT was born at the onset of hyper-accelerating technology. Its mission is to build a platform that navigates the relentless pace of change for Trusted Advisors and their customers. This enables business transformation and establishes an effective, efficient ecosystem for Trusted Advisors to leverage. With those goals in mind, AVANT works with partners to solve complex technology challenges with market-leading solutions. 

The company will offer Aptum’s Advisory & Consulting ServicesPrivate Cloud Managed Services, and Public Cloud Managed Services; plus Application Modernization & DevOps, all to bring best-in-class solutions to their team and customers. 

The agreement is effective immediately, with Aptum solutions now available to AVANT’s Trusted Advisors and their customers in the U.S., Canada and the UK.