エピソード

  • How Cloud Regions Are Auctioning Compute to Fill Gaps

    How Cloud Regions Are Auctioning Compute to Fill Gaps
    2026/06/08
    Episode 38 of The Cloud Business Podcast. Lucas and Luna explore how cloud providers are increasingly auctioning off unused compute capacity through spot and reserved instances, creating a secondary market that enterprises can exploit. They focus on AWS's spot instance market, which now handles over 2.5 million instances daily, and how savvy companies use bidding strategies to cut costs by up to 90%. The hosts discuss real-world examples: a fintech firm running GPU-intensive simulations on spot instances for 80% less, and a media company shifting batch processing to Azure's low-priority VMs. They also unpack the risks, including sudden termination and the need for fault-tolerant architecture. The conversation covers how this auction model is reshaping enterprise cloud strategy, with some companies building hybrid approaches that mix on-demand, reserved, and spot capacity. Lunar notes the growing role of third-party brokers and automated tools that optimize bidding in real-time. The episode closes with a look at how Google Cloud's preemptible VMs and Oracle's bare metal instances are adapting to compete. A brief donation segment highlights listener support via buy me a coffee dot com slash fexingo. #AWS #Azure #GoogleCloud #SpotInstances #CloudAuctions #ComputeOptimization #EnterpriseCloud #CloudEconomics #Fintech #BatchProcessing #GPUComputing #CloudArchitecture #FaultTolerance #CloudStrategy #Business #TechnologyPodcast #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    9 分
  • How Cloud Regions Are Bidding for Enterprise Data Residency

    How Cloud Regions Are Bidding for Enterprise Data Residency
    2026/06/07
    Episode 37 of The Cloud Business Podcast with Fexingo. Lucas and Luna examine how cloud providers are competing for enterprise workloads by offering data residency guarantees beyond compliance checkboxes. They break down AWS's new 'Data Zone' pricing model in Frankfurt, Azure's sovereign-landing-zone strategy in Switzerland, and Google Cloud's 'Assured Workloads' expansion into Japan. The hosts explain why data residency is becoming a negotiable line item in cloud contracts, with specific numbers on cost premiums, vendor lock-in trade-offs, and the emerging role of third-party auditors who certify 'regional isolation' for regulated industries like banking and healthcare. No ads. The show is supported by listeners at buy me a coffee dot com slash fexingo. Fresh angle not covered in prior episodes: the bidding war over where data physically lives, and how enterprises are using it to renegotiate terms. #CloudRegions #DataResidency #AWS #Azure #GoogleCloud #EnterpriseIT #Compliance #SovereignCloud #DataZone #AssuredWorkloads #SwissBanking #GDPR #CloudContracts #RegulatedIndustries #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #CloudInfrastructure Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    12 分
  • How Cloud Regions Are Auctioning Compute to Fill Gaps

    How Cloud Regions Are Auctioning Compute to Fill Gaps
    2026/06/07
    Cloud providers are increasingly using auction-based pricing for spare compute capacity — think AWS Spot Instances on steroids, but now for GPUs and reserved blocks. Lucas and Luna break down how this market works using the example of a mid-size AI startup that saved 62% on inference costs by bidding on idle A100 clusters across three regions. They explore why Google Cloud's dynamic pricing model differs from Azure's quota-based approach, the rise of third-party brokers like Spot by NetApp (formerly Spot.io), and what happens when the auction clears at zero bids. If you manage a cloud budget, this episode gives you a concrete playbook for when to bid versus when to commit. #CloudCompute #SpotInstances #GPUAuction #AWS #Azure #GoogleCloud #EnterpriseInfrastructure #CloudCostOptimization #A100 #Inference #CloudBroker #NetApp #DynamicPricing #BiddingStrategy #ReservedInstances #Business #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    6 分
  • Why Enterprise Cloud Contracts Now Have GPU Resale Clauses

    Why Enterprise Cloud Contracts Now Have GPU Resale Clauses
    2026/06/06
    Episode 35 of The Cloud Business Podcast dives into a new clause appearing in enterprise cloud contracts: the GPU resale right. Lucas and Luna explore why cloud providers are offering customers the ability to resell unused GPU capacity on secondary markets, how this changes the economics of large-scale AI training, and what it means for procurement. With GPU supply still constrained and prices volatile, these clauses could reshape enterprise negotiation leverage. The hosts examine a specific case: a multinational pharmaceutical company that used its GPU resale clause to reduce net compute costs by 18% in Q1 2026. They also discuss the risks, including pricing benchmarks and legal complexity. If you're negotiating a cloud contract for AI workloads, this episode explains a provision that could save millions. #GPUResaleClauses #EnterpriseCloud #CloudContracts #AITraining #CloudProcurement #GPUConstraints #CloudEconomics #AWS #Azure #GCP #CloudNegotiation #SecondaryMarket #AIInfrastructure #CloudCostOptimization #Business #Technology #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    9 分
  • How Cloud Regions Are Competing for Enterprise AI Workloads

    How Cloud Regions Are Competing for Enterprise AI Workloads
    2026/06/06
    In this episode of The Cloud Business Podcast, Lucas and Luna explore how major cloud providers are shifting their geographic expansion strategies to compete on AI-specific factors like inference latency, GPU availability, and data residency. Using concrete examples from AWS, Azure, and GCP's recent region openings, they drill into why enterprise architects should now consider region-level GPU capacity commitments as a negotiation lever in cloud contracts. The hosts examine the emerging practice of 'cloud region arbitrage'—where companies deliberately distribute AI workloads across multiple regions to optimize for cost, latency, and regulatory compliance. They also discuss how secondary regions in places like Malaysia, Spain, and Saudi Arabia are becoming AI hubs due to aggressive tax incentives and renewable energy credits. This episode includes a light listener-support segment. #CloudRegions #AIWorkloads #AWS #Azure #GCP #InferenceLatency #GPUAvailability #DataResidency #CloudRegionArbitrage #EnterpriseAI #CloudInfrastructure #HybridCloud #Multicloud #CloudCostOptimization #Business #Technology #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    10 分
  • Why Cloud Costs Surge When Data Gets Hot

    Why Cloud Costs Surge When Data Gets Hot
    2026/06/05
    Lucas and Luna dig into a cost trap most enterprises overlook: the data temperature problem. Lucas explains how storage tiers — hot, cool, cold, and archive — create explosive cost curves when access patterns shift. He cites a 2025 IDC report showing that 35% of enterprise cloud storage spend is wasted on data stored in hot tiers that is accessed less than once a quarter. They discuss real-world examples from a Fortune 500 retailer that moved cold logs to cold tier and cut storage costs by 68 percent. The hosts also cover how AI data pipelines are making temperature management harder, with GPUs demanding hot access to training datasets while inference data can sit colder. Lucas walks through a simple tagging strategy using AWS S3 lifecycle policies and Azure Blob Storage access tiers. The episode closes with a practical take: enterprises should audit their storage classes every quarter, not just at migration. #CloudStorage #DataTemperature #HotData #ColdData #CloudCosts #AWS #Azure #GCP #IDC #Fortune500 #StorageTiers #LifecyclePolicies #S3 #BlobStorage #AI #GPUs #BusinessAndTechnology #FexingoBusiness Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    6 分
  • Why Enterprise Cloud Contracts Now Include Data Egress Credits

    Why Enterprise Cloud Contracts Now Include Data Egress Credits
    2026/06/05
    In this episode, Lucas and Luna examine the emerging trend of data egress credits in enterprise cloud contracts. They break down why major cloud providers like AWS, Microsoft Azure, and Google Cloud are beginning to offer egress fee waivers or credits for large-scale data transfers, particularly for AI training pipelines and multi-cloud architectures. Lucas shares specific examples, such as Google Cloud's revised egress pricing for customers using third-party data centers and AWS's recent introduction of the 'Data Transfer Out Credit' for select enterprise agreements. Luna challenges whether these credits are genuine cost-saving tools or just marketing gestures, pointing out the fine print around volume thresholds and eligible services. The hosts also discuss how enterprises can negotiate egress credits directly into their contracts, using concrete benchmarks like transfer volumes of 10 terabytes or more per month. The episode closes with a reflection on how the cloud market's maturity is shifting leverage toward buyers, especially those with predictable, high-volume data flows. #CloudComputing #DataEgress #AWS #Azure #GoogleCloud #EnterpriseIT #CloudContracts #CostOptimization #MultiCloud #AITraining #CloudPricing #DataTransfer #Negotiation #Business #BusinessPodcast #FexingoBusiness #CloudStrategy #TechPolicy Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    9 分
  • How Cloud Regions Are Competing on Latency Not Just Price

    How Cloud Regions Are Competing on Latency Not Just Price
    2026/06/04
    In this episode of The Cloud Business Podcast, Lucas and Luna explore how cloud providers are shifting their competitive strategy from pure price wars to latency and performance differentiation. They drill into the specific case of AWS's new edge locations in secondary cities like Columbus, Ohio and Omaha, Nebraska, announced in late 2025, and how these reduce round-trip time by up to 30 milliseconds for midwest enterprises. The hosts discuss why Google Cloud's 'network-first' architecture and Azure's 'express routing' are now being marketed as core advantages, not afterthoughts. They analyze real numbers: a fintech firm in Des Moines cutting app response time by 40% by switching primary regions from us-east-1 to us-east-2 with local edge caching. The conversation also covers the hidden cost of latency in AI inference workloads, where every 10 milliseconds matters for real-time applications. Lucas and Luna examine what this means for enterprise cloud strategy in mid-2026, including how contract negotiations now include latency SLAs that can force providers to refund credits if performance drops below 99.5th percentile thresholds. Tune in for a focused look at why speed is becoming the new battleground in cloud infrastructure. #CloudInfrastructure #Latency #AWS #Azure #GoogleCloud #EdgeComputing #AIInference #EnterpriseIT #NetworkPerformance #CloudRegions #CloudCompetition #SLA #Performance #DataCenter #CloudStrategy #Business #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    13 分