Oskar Ablimit

VMware vCloud Foundation 9 – Licensing Part 1. VCF 9 adopts a streamlined, subscription-based licensing model that simplifies management and compliance. Let's dive deeper into it. – #vExpert Roshan Chavan https://dy.si/P4cZL

4

25/05/2025 - Daily GPU Cloud pricing update - Major hyperscalers align H200 pricing as competition heats up Significant pricing movements announced for tomorrow: H200 pricing alignment: Google Cloud maintains leadership with spot pricing at $3.72/GPU/hour Microsoft Azure reduces 8xH200 pricing by 1.9% to $84.80/hour ($10.60/GPU) Oracle Cloud adjusts to $80.00/hour ($10.00/GPU) for bare-metal Market stabilizing around $10-11/GPU/hour for on-demand A100 series adjustments: Hyperstack cuts A100 PCIe costs by 3.6% to $1.35/GPU/hour RunPod reduces A100 SXM by 3.1% to $1.89/GPU/hour Gcore optimizes 8xA100 SXM pricing by 4.6% to $1.67/GPU/hour Legacy GPU optimization: CUDO Compute reduces V100 pricing by 10.3% to $0.35/GPU/hour goo Market insights: H200 pricing war intensifies among major providers Growing gap between spot and on-demand pricing A100 remains competitive for cost-sensitive workloads Legacy GPU prices continue downward trend -- You are a GPU Cloud consumer? Sign In to get the latest opportunities pricing: https://gpucompare.com/ You are a GPU Cloud provider? Sign In to claim your Page and share opportunities: https://lnkd.in/gVfSPjUe

4

Nix users: Meet your first and only Enterprise Substituter. ❄️+🐸 Native Nix support has officially landed in JFrog Artifactory! We are proud to be the exclusive enterprise-grade provider offering a universal binary cache for Nix. No more bandwidth bottlenecks, no more dependency fragility, and no more "workarounds" to get Nix into your secure pipeline. Why it’s a game-changer: ‣ Exclusive native support: The only platform to manage #Nix alongside #Docker, #npm, and #Maven. ‣ Survival mode: Cache cache.nixos.org locally to survive upstream outages. ‣ Private & Secure: Host proprietary Nix builds behind a true enterprise firewall. Stop rebuilding. Start substituting. Learn more: https://bit.ly/4rBOWcU

35

1 Comment

Your cloud supplier just grew 48%. Did you? FY 2025 Hyperscaler numbers: Provider Q4 Cloud Revenue YoY Growth ✅ AWS $35.6bn 24% ✅ MS Azure $32.9bn 29% ✅ ✅ GCP $17.7bn 48% Now compare that with: G7 enterprise revenue growth? 1–5%. FTSE 100? Mid-single digits. So the maths is simple. If your revenue grows at 3% And your cloud supplier grows at 30% Cloud becomes a bigger % of your P&L. Which means one thing: 📈 Unless productivity rises equally fast, 📉 EBITDA compresses. 🛑 Quietly. 🛑 Gradually. 🛑 Compounded. Example. £50m revenue. 20% EBITDA = £10m. If £1m of cloud spend is 10% inefficient, £100k disappears. That’s 0.2 margin points gone. Now compound that... This isn’t anti-hyperscaler. They’re executing brilliantly. It’s a governance issue. In 2026, cloud optimisation isn’t a tech decision. It’s margin discipline. So here’s the question: When your cloud supplier grows 10x faster than you do… Is that strategic scaling? Or silent EBITDA erosion? #CloudEconomics #EBITDA #CapitalAllocation #CFO #BoardLevel #FinOps #MarginDiscipline #TechStrategy

16

4 Comments

Both OCI File Storage (FSS) and Lustre drastically reduce container image fetch and execution times on repeat accesses. Learn more: https://lnkd.in/gYu2nMJV

2

📣 BIG AWS re:Invent news: Wiz Exposure Management is officially GA. UVM + ASM in one platform - built to cut the noise and focus on real risk. With Wiz Exposure Management you can ↓ 1) Unify all risk data across cloud, code, on-prem, and hybrid. 2) Validate what's actually exploitable from an attacker's view. 3) Prioritize with the Wiz Security Graph and see real attack paths. 4) Fix faster with ownership mapping, smart assignments, and AI-powered remediation. It's officially time to reduce exposure - not alerts. 🔗 Learn more >> https://lnkd.in/eRhR65Pi

3

🚀 Okta Application Onboarding & SSO — Connecting the Dots! As part of my Okta deep dive, I’ve been geeking out over application onboarding, SSO protocols, and the magic of provisioning. 😉 Here’s what I explored: ✅ What is Application Onboarding? It’s the process of integrating new apps into Okta, so users can access them securely — without IT pulling their hair out. 🌟 ✅ Provisioning vs. De-provisioning: Provisioning: Automatically creates, updates, or assigns users to apps. De-provisioning: Instantly cuts off access when someone leaves — hello, security hygiene! ✅ How Do We Integrate Applications? Use Okta’s Application Catalog (it’s like the app store for identity!). Choose the right SSO protocol: 🔹 SWA → Secure Web Authentication 🔹 SAML → Security Assertion Markup Language 🔹 OIDC → OpenID Connect ✅ SSO vs. Identity Federation: SSO: One login, multiple apps within your org. Federation: Linking identity across orgs or domains (think partners, B2B). ✅ How the Protocols Work (and Where to Use Them): SWA: Credential vaulting. Works with legacy apps. Example: Zoom. SAML: Passes signed assertions (user ID, attributes) between IdP and app. Example: Dropbox. OIDC: Built on OAuth 2.0, issues ID, access, and refresh tokens. Example: ClicData. 💥 Why SSO Matters: It reduces password fatigue, improves security, and boosts productivity. Users love it. Security teams love it even more. ❤️ 🛠️ Example Applications I Explored: SWA: Zoom, Zoom (custom) SAML: Dropbox, Dropbox (custom) OIDC: ClicData, ClicData (custom) ✨ Takeaway: Okta isn’t just an SSO tool — it’s your control center for secure, scalable, automated access across all applications. Blog Link : https://lnkd.in/gsXqGEqV #Okta #IAM #SSO #Provisioning #SAML #OIDC #SWA #IdentityManagement #CyberSecurity #TechWriting #OktaOnboarding #OktaIntegration

9

📊 Optimize your cloud data protection costs without compromising security. Keith Sng from Veeam and Desmond Lai from AWS break down: - How Veeam Backup for AWS uses snapshots, S3, and Glacier - Cost-saving strategies for backup storage - How retention & data change rates affect your budget - Whether you’re migrating workloads or refining your cloud strategy, this session gives you the insights to make smarter, cost-aware decisions. Watch now and start saving

2

🚀 Exciting Case Study: Democratizing AI Implementation at Musinsa I'm thrilled to share how our AWS team Minki Chae at Musinsa, Korea's largest fashion platform with 15M+ members, successfully implemented an AI-based product recommendation and search system. 🔍 The Challenge: We needed to enhance our ability to recommend similar products based on influencer-worn items in our SNAP service. Traditional keyword-based systems weren't cutting it anymore. 💡 The Solution: We built an end-to-end AI system using: - YOLOS-Fashionpedia for object detection - FashionCLIP 2.0 for image/text embedding - AWS services (SageMaker, OpenSearch, Lambda, Bedrock) 📊 The Results: - 72% precision for top-1 results - 81% precision for top-5 results - 0.76 Mean Reciprocal Rank 🌟 What makes this truly special? The entire project was executed by two backend developers with NO prior AI experience. This breakthrough proves that AI implementation doesn't have to be gatekept by specialized AI teams. 🎯 The Impact: This success has transformed how we approach AI projects at Musinsa. Teams can now independently pursue AI initiatives without waiting for central AI team bandwidth. We've effectively democratized AI development within our organization! 💪 Key Takeaway: Don't let lack of AI expertise hold you back. With the right tools and determination, development teams can successfully implement AI solutions independently. What are your thoughts on democratizing AI implementation within organizations? #AI #SoftwareDevelopment #Innovation #TechTransformation #AWS #Fashion #Engineering https://lnkd.in/gzec7bvX

5