《Omdia&谷歌云：2024年邊緣計算狀況報告：泛在AI賦能行業解鎖現代化商業用例（英文版）（28頁）.pdf》由會員分享，可在線閱讀，更多相關《Omdia&谷歌云：2024年邊緣計算狀況報告：泛在AI賦能行業解鎖現代化商業用例（英文版）（28頁）.pdf（28頁珍藏版）》請在三個皮匠報告上搜索。

1、COMMISSIONED BYRESEARCHED BY2024 State of Edge Computing:How industries are leveraging AI anywhere to unlock modern business cases CATALYST Edge computing is not new but has taken on a new level of importance with digitization of organizations and consumers,increasing demands to process data locally

2、 on-premises,and leveraging the latest AI and cloud-based innovation.IT decision makers and buyers are exploring new use cases,and their vendors are developing software and hardware to meet new requirements for edge located equipment.Several industry verticals have been leveraging edge computing for

3、 years to transform their operations,with manufacturing using it for automation,while others,such as municipal governments,are just starting smart city project rollouts.It is evident that we are only at the beginning of exploring various use cases.As we complete the deployment of known applications,

4、many additional use cases will undoubtedly emerge.OMDIA VIEWThe increasing prominence of AI and Generative AI(GenAI)is anticipated to profoundly transform business and the nature of work.This expectation resonates across all leadership levels,from the C-suite overseeing digital transformation initia

5、tives to practitioners actively testing,executing,and scaling cutting-edge innovations to deliver value.GenAI will spawn a whole new approach to thinking about the nature of work and who and where it should be performed.GenAI is poised to unlock new skillsets and present novel challenges,transformin

6、g various aspects of the business landscape through elements such as creative problem solving,data interpretation and analysis,and technical proficiency in training and fine-tuning GenAI models.This will lead to the development of new devices and software technologies to improve business processes,r

7、elieve humans of repeatable tasks,and make life more fun.AN OMDIA EBOOK2024 STATE OF EDGE COMPUTING02Summary?HOMEAt the same time,the speed and effectiveness of delivering value to customers and staff locally requires compute processing on-premises where the collection and real-time processing of da

8、ta are becoming increasingly important.As a result,latency and bandwidth are becoming key performance determinants and are driving the need for better telecommunication networks and more computing power to be placed closer to end users and machines.Security and data volume are also factors that can

9、influence end users to place more compute at the edge.Enterprises,telecommunications network providers,and cloud service providers(CSP)have adopted edge computing strategies.For example,CSP approaches are based on the premise of deploying hardware to the edge that is connected to and compatible with

10、 the core cloud solution offered by the cloud provider.Many enterprises have opted to own their own hardware at the edge instead of relying on CSP-or telco-operated data centers(DCs).In fact,enterprises were early adopters of edge computing.Many enterprises have a distributed business models where a

11、pplication support is required at multiple branches,offices,or stores.Additionally,many enterprises have been running latency-sensitive workloads such as healthcare and industrial applications;on-site data consolidation,data sharing,and analytics;and retail store management.With GenAI the developmen

12、t of the applications requires in some cases,large amounts of compute resources such as GPUs and TPUs.However,the execution of these applications is likely to be distributed to the edge for privacy and latency reasons.On-premises computing is well-established,but the integration of cutting-edge clou

13、d infrastructure and AI represents a transformative shift.Cloud technologies are empowering developers to create agile and responsive applications tailored to evolving customer needs,both internal and external.Additionally,the swift advancements in AI and Generative AI are unlocking new opportunitie

14、s to address complex,costly,and resource-intensive challenges,such as predictive analytics and visual inspection.The convergence of on-premises computing,cloud technology,and AI heralds a new era for IT decision-makers:AI Anywhere.KEY MESSAGESAN OMDIA EBOOK032024 STATE OF EDGE COMPUTINGAdoption of e

15、dge is evolving driven by the need for low latency,security,and data volume requirements.AI anywhere is a key driver supporting mission critical use cases requiring an inclusive edge,AI,and cloud strategy.Industry View:Industries are unlocking new use cases to improve business analytics,security,and

16、 operations.Cloud and open ecosystems unlock developer agility to build,deploy,and scale applications at the edge.Spend and scale is increasing in coverage across industries.?HOME100%OF RESPONDENTS PLAN TO USE EDGE COMPUTING IN THE NEXT 12 MONTHSFigure 1 shows that not all use cases are suitable for

17、 the public cloud.Consider the following scenarios:z Computational Requirements:When processing must continue uninterrupted,even if the connection is down.z Data Gravity and Volume:When handling high-volume,noisy,or costly data that is time-consuming to transmit elsewhere.z Latency and Speed:When qu

18、ick decision-making is critical,windows of opportunity are narrow,or local interactions are required.z Regulated Data:When regulatory measures and policies mandate that data must remain on-premises.The survey results,as shown in Figure 2,clearly indicate that edge computing is a priority for all org

19、anizations.Edge computing is being integrated into various transformation initiatives,ranging from enterprise-wide digital transformations to more targeted network transformation projects.Notably,14%of respondents reported not using the cloud,with the healthcare and life sciences sectors leading at

20、25%.AN OMDIA EBOOK042024 STATE OF EDGE COMPUTINGAdoption of edge is evolvingLatency-sensitive applicationsComputational requirementsSecurity requirementsData gravity,data volume0%5%10%15%20%25%30%35%33%18%29%20%Figure 1:Top reasons for deploying edgeS7:WHICH WOULD YOU RATE AS THE TOP REASONS FOR EDG

21、E COMPUTING DEPLOYMENT?SOURCE:OMDIA?HOMEThis underscores that not all use cases are suitable for the cloud.Figure 3 reveals plans to significantly increase the number of edge locations by 2026,with the most substantial growth in deployments numbering in the hundreds.Specifically,there is a 15-percen

22、tage point increase in these deployments from 2024 to 2026.Further analysis shows that enterprise organizations(those with revenues greater than$1Bn)will invest more in edge.The survey showed 40%of enterprises($1Bn plus)will invest more than$500M compared to 9%of mid-market and SMB(less than$1Bn org

23、anizations).This data correlates with the IT budget analysis Omdia performs annually.In Omdias IT budget and spending forecast the distribution of IT budget spending by subcategory shows a consistent trend across all organization sizes.However,for large organizations it shows that by 2027,the averag

24、e IT spend per year on infrastructure and applications will be:applications$310 million and infrastructure$270 million.Customer services and technology and process are the clear top two departments looking to use edge.The survey identified that when it comes to how and for what the edge will be used

25、,latency sensitivity and security are the top two uses.Latency is particularly important as with the GenAI revolution,many of the use-cases for GenAI are likely to be placed so that the customer experience and privacy are paramount.This means that these inferencing use cases will be sensitive to lat

26、ency and privacy,so are prime candidates for edge deployments.Beyond the initial customer focused use cases,organizations consider the ability to be predictive and obtain real-time risk analysis as the most transformative capabilities the edge and AI can deliver.Organizations expectations from edge

27、show it must deliver improved security and compliance.AN OMDIA EBOOK052024 STATE OF EDGE COMPUTING0%10%20%30%40%My organization has no plans to adopt edge computing0%Edge computing is part of an enterprise-wide digital transformation initiative36%Edge computing is part of a network transformation in

28、itiative 35%Edge computing is part of a standalone project for specific line of business30%0%10%20%30%50%40%60%8%5%1%0%42%15%32%4%47%58%44%59%3%21%20%32%0%1%3%5%1 to 910 to 4950 to 99Thousands(1000 or more)Hundreds(100 to 999)Figure 2:Business plans for edge in 2024S11:WHICH STATEMENT BEST DESCRIBES

29、 HOW EDGE COMPUTING IS PLANNED IN YOUR ORGANIZATION?SOURCE:OMDIAFigure 3:Number of edge deployments 2024 to 2026Q19:HOW MANY EDGE LOCATIONS DO YOU PLAN TO DEPLOY COMPUTING INFRASTRUCTURE TO NOW AND IN THE FUTURE?2024 Mid-market 2026 Mid-market 2024 Enterprise 2026 EnterpriseSOURCE:OMDIA?HOMEAI anywh

30、ere is a key driver supporting mission critical use cases requiring an inclusive edge,AI,and cloud strategyAN OMDIA EBOOK06Figure 4 illustrates the adoption of AI across various industries,highlighting both early adopters and more cautious entrants.The banking sector has been proactive,deploying AI

31、extensively as a strategic measure to enhance security in an era where deep fakes are increasingly used for fraud.Conversely,the utilities sector is more reserved in its AI adoption due to its classification as a critical infrastructure industry,facing unique challenges.However,as a highly distribut

32、ed and infrastructure-intensive industry,utilities will increasingly leverage edge computing and AI to drive business transformation.This contrast is evident in the projected edge deployments over the next two years.By 2026,38%of banking institutions plan to deploy AI to hundreds of edge locations,w

33、hile the majority of utilities(71%)will limit deployments to fewer than a hundred locations.2024 STATE OF EDGE COMPUTINGBankingHealthcareManufacturingPublic sectorRetailTelecommunicationsUtilities0%5%10%15%20%25%22%12%15%12%10%16%11%15%16%14%16%13%10%17%Figure 4:The AI journey by industryQ8:WHAT IS

34、THE STATE OF AI DEPLOYMENT AND PILOT PROJECTS AT YOUR COMPANY TODAY?Deploying AI Piloting AI SOURCE:OMDIA?HOMEAN OMDIA EBOOK072024 STATE OF EDGE COMPUTINGFigure 5:Vertical deployment of AIQ10:IN WHICH CORE BUSINESS DEPARTMENT ARE YOU DEPLOYING AI?Banking Healthcare Manufacturing Public sector Retail

35、 Telco Utilities SOURCE:OMDIAStrategic managementTechnology and process developmentHuman resource managementGeneral managementCustomer and after sales serviceMarketing,sales,and customer accountsProduct or service deploymentOperationsProcurement,logistics and distribution0%50%100%150%200%250%300%350

36、%400%450%30%19%62%32%37%42%51%23%21%41%60%25%23%36%36%16%11%13%11%6%5%27%39%58%47%23%63%35%39%58%47%34%63%22%22%42%37%34%19%43%43%68%61%51%58%35%47%5%29%15%30%37%52%18%40%42%57%57%68%37%57%79%Deployment trends also vary by organizational size.Enterprise organizations($1Bn-$5Bn annual revenue)are lea

37、ding in strategic AI management deployment,with 43%incorporating AI strategically.In comparison,only 7%of small organizations(less than$499M annual revenue),20%of mid-sized organizations($500M-$999M annual revenue),and 27%of large enterprises(greater than$5Bn annual revenue)are doing the same.The pr

38、imary use cases for AI deployment are technology and process development,followed by customer service,as shown in Figure 5.Enterprise organizations($1Bn-$5Bn annual revenue)are at the forefront of technology and process development(58%),with customer and after-sales services(52%)and operations(44%).

39、These leading use cases are predominantly driven by data gravity and security concerns,with 61%of respondents deploying AI for these purposes,compared to 42%for computational needs and 54%for latency considerations.While these early use cases set the foundation,each industry will continue to develop

40、 its unique agenda for integrating edge computing and AI.?HOMEAN OMDIA EBOOK082024 STATE OF EDGE COMPUTINGREALITY VS DREAM:PARTNERS ARE KEY TO EDGE AND AI.Figure 6 illustrates that the majority of edge computing spending by both enterprise and mid-sized organizations is directed towards cloud partne

41、rs.However,mid-sized organizations tend to engage more with OEMs,system integrators,and MSPs compared to their enterprise counterparts.This distribution is influenced by the existing relationships these organizations have with their partners.Mid-sized and SMB organizations often rely on local or reg

42、ional partners due to their limited resources,which makes it difficult to fully leverage the self-service model commonly offered by cloud partners.In contrast,enterprise organizations,with their in-house expertise and knowledge of cloud technologies,are more inclined to utilize cloud partners for ed

43、ge-related projects.For AI deployments enterprise organizations are evenly split between in-house software development on cloud-based AI infrastructure(35%)or partner-led on cloud(34%),while 20%are completely in house.Conversely,mid-sized organizations are more likely to use partner-led in cloud(42%

44、).The difference with AI compared to edge is that AI is a new technology and as such skills and knowledge on the topic are difficult to find,and many smaller regional partners are only now just building out their AI practices,while cloud partners have more mature practices in AI.Interestingly data g

45、ravity(61%)use cases are more likely to be partner-led than security(47%).This seems at odds with Omdias observations where the continued evolution of Secure Access Service Edge(SASE)continues to see growth in the partner network,where vendors continue to fuel their SASE initiatives with acquisition

46、s,funding,and partnerships.When it comes to expectations of partners delivering AI and edge,the ability to guarantee service level performance is the least expected partner expertise.This is not unexpected as the market is at a very early stage of development and most partners lack the experience to

47、 be able to provide such guarantees.All industry verticals expect partners to have GenAI capabilities,with public sector(73%)the industry most expectant compared to healthcare and telco who are the least expectant(60%).Cloud and open ecosystems unlock developer agility to build,deploy,and scale appl

48、ications at the edge.Figure 6:Where is the spending being directed for edge?Q15:WHICH PARTNERS ARE YOU WORKING WITH/ARE YOU MOST LIKELY TO WORK WITH FOR EDGE COMPUTING?Mid-Market Enterprise SOURCE:OMDIA0%5%10%15%20%25%30%Cloud service providerSystem integratormanaged service providerTelecommunicatio

49、ns service providerIT equipment vendor(OEM)Software vendor24%27%15%12%19%19%15%17%16%14%10%11%1%0%Others?HOMEAN OMDIA EBOOK092024 STATE OF EDGE COMPUTINGFigure 7 highlights cybersecurity and data protection risks as the top barriers to faster edge adoption across regions,with 41%in EMEA,47%in North

50、America,and 46%in APAC citing these concerns.While cybersecurity poses a slightly greater issue for mid-sized and SMB organizations,the lack of skills is a more significant barrier for enterprises.Interestingly,12%of respondents in EMEA perceive no barriers to edge adoption.Regarding partner experti

51、se,North America shows the least concern about the absence of edge services among partners,likely due to the abundance of partners in the region.In contrast,other regions face challenges securing partners with edge capabilities due to their limited availability.This is gradually changing as partners

52、 expand their edge and AI service portfolios.For example,17%of retail organizations view the lack of key technology partners with edge services as a barrier,ranking it as the seventh most important barrier for retail,compared to the lowest rank(11th)for other industries,except utilities.This suggest

53、s a scarcity of partners with edge services in the retail sector or a higher demand for edge deployments in retail.Currently,47%of retail respondents have fewer than 49 edge deployments,but by 2026,only 8%will remain at this level,indicating that 92%of retail organizations will have 50 or more edge

54、locations.This increased demand and the current lack of partner services is likely to drive the retail sectors heightened sensitivity to partner capabilities in edge services.Figure 7:Barriers to faster edge deploymentQ21:IN YOUR VIEW,WHAT ARE THE BIGGEST BARRIERS TO THE UPTAKE OF EDGE COMPUTING IN

55、YOUR ORGANIZATION?Mid-Market Enterprise SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)0%10%5%15%20%25%30%35%45%40%50%Lack of internal skills28%36%Lack of compelling use case22%14%Lack of low-latency networks25%22%Immature technology standards29%22%Cybersecurity47%42%Perceived high costs

56、32%34%Management/orchestration16%11%IT and OT team silos25%24%Fear of vendor lock-in16%14%Partner has no edge services12%10%Others1%1%There are no barriers8%12%?HOMEAN OMDIA EBOOK102024 STATE OF EDGE COMPUTINGSpend and scale is increasing in coverage across industriesENTERPRISES ARE SPENDING ON AVER

57、AGE$740K ON EDGE COMPUTING THROUGH 2026Of the 365 enterprises generating more than$1B in revenue that we surveyed,edge spending is an extension or part of the existing cloud budget in most organizations,with 32%of enterprise organizations classifying it this way,a sizeable minority,25%,of enterprise

58、 organizations classifying edge spending as part of digital transformation budget.In mid-market and below the difference is less pronounced,with 28%of edge spending funded from the cloud budget,and 25%from the digital transformation budget.Figure 8 shows enterprise organizations($1Bn plus)will be sp

59、ending more than twice that of SMB and mid-size organizations by 2026 on edge.Enterprise spend on average will be$990K compared to$410K for mid-market and below.Value generation from edge is mostly expected to be derived from cost savings with 67%of all organizations expecting greater than 11%in cos

60、t savings from edge deployments compared to 60%that expect less than 10%revenue generation from edge.Omdia considers that this perspective is driven mostly by the currently identified use cases,see Figure 9 for a per industry perspective.These early use cases are nearly all related to either improvi

61、ng existing processes or providing better insights into existing data assets.Beyond this first wave of edge and AI deployments,the second phase will be more diverse and focused on new business opportunities such as those being pursued by the telcos.Figure 8:Investment in edgeQ18:HOW MUCH DO YOU EXPE

62、CT TO INVEST/HAVE INVESTED IN EDGE COMPUTING HARDWARE,SOFTWARE,MANAGED AND PROFESSIONAL SERVICES DURING THE INITIAL 24 MONTHS OF DEPLOYMENT?Less than$1Bn$1Bn plus SOURCE:OMDIA0%10%20%30%50%40%60%$100,000 to$499,99951%42%$500,000 to$999,99932%8%$1 millionto$4.9 million1%7%$5 millionto$9.9 million0%0.

63、5%$10 million or more0%0.3%$50,000 to$99,99930%16%$10,0000.7%0.3%$10,000 to$49,9999%2%?HOMEAN OMDIA EBOOK112024 STATE OF EDGE COMPUTINGFigure 9:Investment in edge by industryQ18:HOW MUCH DO YOU EXPECT TO INVEST/HAVE INVESTED IN EDGE COMPUTING HARDWARE,SOFTWARE,MANAGED AND PROFESSIONAL SERVICES DURIN

64、G THE INITIAL 24 MONTHS OF DEPLOYMENT?$10,000$10,000 to$49,999$50,000 to$99,999$100,000 to$499,999$500,000 to$999,999$1 million to$4.9 million$5 million to$9.9 million$10 million or more SOURCE:OMDIA0%10%20%30%50%40%60%48%10%2%4%21%16%0%0%45%6%0%5%18%23%1%1%44%7%0%3%25%21%0%0%37%3%0%2%33%25%0%0%47%3

65、%0%5%20%25%0%0%53%3%0%4%21%18%1%0%48%3%1%5%16%27%0%0%UtilitiesBankingHealthcareManufacturingPublic sectorRetailTelco?HOMEAN OMDIA EBOOK122024 STATE OF EDGE COMPUTINGFigure 9 shows the expected spend by industry on edge over the next 24 months for organizations of all sizes.Omdia has estimated that b

66、anking and finance will spend the most with an average of$670K,compared to utilities who expect to spend the least with an average of$342K.Other key spending indicators for edge are:56%of large($5Bn plus)enterprises will spend between$0.5-$5M on edge.7.5%of banking organizations expect to invest mor

67、e than$1M in edge compared to 2.1%in manufacturing.42%of enterprises with more than$1B plan to spend$100,000 to$500,000 on edge33%of retail organizations expect to spend more than$500,000 in edge in the next 24 months.7.2%of organizations in APAC expect to invest more than$1M in edge compared to 4.7

68、%in North America and 3.1%in EMEA.?HOMEAN OMDIA EBOOK132024 STATE OF EDGE COMPUTINGExpectation vs Reality:industries are unlocking new use cases to improve business analytics,security,and operations.RETAIL:MORE THAN 44%PRIORITIZE FRAUD LOSS PREVENTION AND PERSONALIZATION.Figure 10 shows the response

69、 of retail organizations to the survey in terms of the use cases being considered and deployed.Physical security was the top use case,which for retail is all about loss prevention and fraud detection.Improving customer experience with personalization and in-store promotions was the second use case,p

70、articularly for on-line retailers,which agreed with Omdias IT Enterprise Insights(ITEI)survey 2023(n=4800)where it was also second to increasing operational efficiency.While this may appear to contradict Figure 10,loss prevention in retail is a major part of operational efficiency.Interestingly,sust

71、ainability is not a driver for retail,Omdias ITEI 2023 survey agrees as sustainability was the second lowest ranked use case.Retail in terms of operational efficiency is also focused on employee productivity and the introduction of technology in store,at the edge,is seen as a big driver of cost savi

72、ngs.In fact,49%of retail decision makers expect to get 11-20%in cost savings from deploying edge,while 47%expect to get 5-10%return in terms of increased revenue from edge.This shows that initially at least,retail see edge as a solution to deal with operational efficiency objectives in store as busi

73、ness analytics at the edge is the most popular workload being deployed,84%.Retail is also more likely to use a cloud partner for deployment of the edge,61%,which is an indication of where the core data needed at the edge is processed.In terms of expected outcomes from edge initiatives,retail decisio

74、n makers put improved security or compliance as the top expectation,with improved compute performance a close second,48%and 44%respectively.Retail decision makers do not expect to use open-source AI as this was the lowest ranked expected outcome,17%.Retail organizations report that 26%are scaling AI

75、 deployments across multiple business functions,with only 6%having no plans for AI.In fact,55%of retail organizations are investigating or piloting AI use cases,which indicates that in retail the use of AI is seen as must have technology.?HOMEAN OMDIA EBOOK142024 STATE OF EDGE COMPUTINGFigure 10:Top

76、 Retail use casesQ1:WHAT ARE YOUR ORGANIZATIONS GREATEST CHALLENGES/AREAS OF IMPROVEMENT(RETAIL,ECOMMERCE,AND CONSUMER PRODUCT GOODS)?SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)Loss prevention and fraud detectionPersonalization and in-store promotionsEmployee productivity,skills,and

77、safetyAugmenting digital experiencesInbound logistic/inventory controlSustainability and environmental proceduresCyber security(e.g.,ransomware,malware)New product/service developmentOutbound logistics/distributionMarketing,sales and customer engagementStore analyticsSmart checkouts and frictionless

78、 shoppingInventory detection and recommendations0%5%10%15%20%25%30%35%40%45%50%47%44%39%38%37%33%32%31%28%27%26%26%19%?HOMEAN OMDIA EBOOK152024 STATE OF EDGE COMPUTINGMANUFACTURING:41%+RANK PREDICTIVE MAINTENANCE AND INVENTORY MANAGEMENTFigure 11 shows factory floor analytics are the top use cases f

79、or manufacturing organizations for edge computing and AI.We have seen similar results from manufacturing companies according to Omdias ITEI 2023 survey,that found the drive for efficiency and reduced down-time is a key priority.The ultimate purpose of this is to support the business priority in manu

80、facturing,which according to Omdias ITEI 2023 survey is to increase revenue and grow the business.Many manufacturing organizations operate a lean model where inventory is delivered just in time for the production need.The survey confirmed this,as asset inventory and tracking is the top workload for

81、edge(61%)for manufacturing organizations.Predictive maintenance is another obvious use case for manufacturing where early detection of potential problems can be accommodated in the production schedule and therefore improve the line efficiency.In Omdias ITEI 2023 survey,intelligent automation and AI

82、in manufacturing was one of the top technology trends for investment in 2024,with over 16%of respondents ranking it as the most important,making it fourth overall.In manufacturing,process optimization is not considered as a priority use case because most manufacturing organizations have developed ma

83、ture,well-defined processes.However,where edge and AI are considered important is in terms of product quality,where manufacturing line monitoring can be used to detect product quality issues and ensure these are remediated quickly,therefore reducing waste.Manufacturing sees edge and AI in equal term

84、s when it comes to financial benefits.The survey found that 42%of manufacturing organizations expect to get more than 11%return in revenue from edge.Meanwhile,58%expect to receive more than 11%in cost savings from edge.Manufacturing represented 15%of the non-cloud users,and this can be seen as only

85、51%of manufacturing organizations leverage a cloud partner to deploy at the edge.Manufacturing organizations report that currently they have identified at least one-use case and are developing a pilot,35%.Manufacturing is less advanced than other industry sectors as only 10%of manufacturing organiza

86、tions are scaling AI deployments across multiple business units.However,manufacturing organizations are not Luddites as only 6%report no plans for AI currently.Omdia considers that while cautious,manufacturing has recognized the value of AI and edge,48%of manufacturing organizations state risk analy

87、sis and mitigation in real-time as the most significant capabilities that will transform its business.?HOMEAN OMDIA EBOOK162024 STATE OF EDGE COMPUTINGFigure 11:Top Manufacturing use casesQ2:WHAT ARE YOUR ORGANIZATIONS GREATEST CHALLENGES/AREAS OF IMPROVEMENT(MANUFACTURING)?SOURCE:OMDIA(RESPONDENTS

88、COULD SELECT MORE THAN ONE OPTION)Inbound logisticsPredictive maintenanceSustainability and environmentalCustomer engagementProduct developmentPhysical securityProduct quality controlOutbound logistics/distributionCyber securityWorkplace safetySupply chain optimizationManufacturing optimizationWorkf

89、orce assistance0%5%10%15%20%25%30%35%40%45%50%45%45%42%41%38%37%34%31%27%26%23%11%8%?HOMEAN OMDIA EBOOK172024 STATE OF EDGE COMPUTINGHEALTHCARE IS ALL ABOUT THE PATIENT EXPERIENCEFigure 12 shows that in healthcare it is all about the patient and how patients can be remotely monitored.The clear top u

90、se case was in-house patient monitoring,which aligns with Omdias ITEI 2023 survey data showing healthcares top business challenges are improving efficiency and improving patient experience.The top workload being deployed by healthcare at the edge is data collection,which is consistent with the use c

91、ase.Another key area for healthcare is concerns over cyber security,not physical security,which Omdia translates to mean the access to the healthcare facility is open,but the IT data and equipment are where the concerns exist.Healthcare is less sure in terms of the cost savings from edge and AI,with

92、 32%expecting a 5-10%cost saving and 32%expecting a 11-20%cost saving.However,they are more consistent in terms of the expected revenue,54%expect less than 10%in terms of new value from edge.This uncertainty in terms of the cost savings from edge is not being seen in the intentions,as healthcare pla

93、n a three-fold increase in the number of edge deployments in the hundreds by 2026 compared to today.Healthcare shows a leading approach to AI and edge,with 21%of respondents reporting a live AI deployment in at least one business function,with 20%of respondents scaling AI across multiple business fu

94、nctions.However,Healthcare does have a slightly higher percentage of those with no plans than other industries,8%.Healthcare organizations are very clear when it comes to their expectations of what is required for a successful AI and edge implementation.91%put the ability to deploy,manage,and scale

95、standardized software from cloud to edge.This clear expectation signals that Healthcare believes standards are needed in the AI and edge market if the technology is going to deliver its expected value.Figure 12:Healthcare use cases for edgeQ3:WHAT ARE YOUR ORGANIZATIONS GREATEST CHALLENGES/AREAS OF

96、IMPROVEMENT(HEALTHCARE AND LIFE SCIENCES)?SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)In-house patient monitoringRemote patient and equipment monitoringQuality controlCyber security(e.g.,ransomware,malware)Supply chain efficiency(e.g.,track biospecimens or equipment)Sustainability and

97、 environmental proceduresPrecision surgeryClinical trials(e.g.,continuous patient data collection)Worker safety and trainingTrack location and status of beds,medical equipment,and staffAdministrative tasksPhysical security and asset protection0%10%20%30%40%50%46%41%40%37%35%33%30%29%27%24%24%23%?HOM

98、EAN OMDIA EBOOK182024 STATE OF EDGE COMPUTINGRegulated industries have different requirements from edgeThe regulated industries(this includes banking,telecom,utilities and public sector)in this survey were all given the same use case questions to answer,and the results are less industry specific tha

99、n the industries previously detailed.BANKING-75%PRIORITIZE PRODUCT/SERVICE INNOVATIONBanking decision makers are prioritizing innovations to improve customer and staff experiences to differentiate themselves in the market.Figure 13 shows that product and service innovation is ranked higher than frau

100、d detection as a use case.Omdia considers that this use case covers emerging aspects in the banking industry.For example,fraudsters are manipulating computer-generated images to animate them with motion,resulting in deepfakes used to defraud victims.These tools utilizing AI are currently evading the

101、 mechanisms that are being deployed to defend against attacks,which means organizations are struggling to adapt to emerging threats,and remote identity verification techniques are fast becoming outdated.Unsurprisingly,the digital identity market is attracting the attention of regulators and market p

102、articipants alike,and although in its infancy,it has the potential to become a major growth area for edge and AI in the banking and finance sector.In fact,45%of banking respondents expect to invest$100K-$500K in edge over the next 2 years with 51%expecting the number of edge locations to be in the 5

103、0-99 number range by 2026.Banking considers edge as a cost saving technology as 44%of banking respondents expect a 11-20%cost saving from edge compared to the majority,53%,that are less optimistic about value generation(less than 10%).?HOMEAN OMDIA EBOOK192024 STATE OF EDGE COMPUTINGFigure 13:Bankin

104、g use cases for edgeQ4:WHAT ARE YOUR ORGANIZATIONS GREATEST CHALLENGES/AREAS OF IMPROVEMENT(BANKING AND FINANCIAL SERVICES)?SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)Product/service innovationFraud preventionMake complex data,intuitively accessible and actionableLarge scale data sea

105、rch,retrieval and analysisContent generation at the click of a buttonMarketing,sales,and customer engagementCyber security(e.g.,ransomware,malware)Improved customer experiencesPhysical security and asset protectionSustainability and environmental proceduresEmployee safety and trainingAutomate incide

106、nt resolutionEmployee productivityRegulatory complianceSupply chain optimization0%10%20%30%40%50%60%70%80%73%50%41%40%40%37%30%29%29%26%25%25%20%18%0%Bankings expectation of what is required to deliver a successful AI and edge deployment show three key capabilities,the ability to deploy,manage,and s

107、cale standard software,security systems and tools upgrade,and location level network speed upgrade,79%,77%,and 75%put as essential respectively.The leading expectation can be linked to the banking organizations stated business objectives for 2024 of delivering increasing operational efficiency.This

108、was the top challenge identified in Omdias IT Enterprise Insights 2023 survey(n=627)with 24%of respondents ranking it number one.?HOMEAN OMDIA EBOOK202024 STATE OF EDGE COMPUTINGPUBLIC SECTOR 60%ARE FOCUSED ON PHYSICAL SECURITY AND ASSET PROTECTIONFigure 14 shows that for the public sector,physical

109、security and asset protection is the top use case.Considering the diverse nature of public sector organizations,this shows a remarkable level of agreement.Omdias IT Enterprise Insights Survey 2023 confirms that government agencies expect modest increases in IT spending to continue,with over 50%of ag

110、encies anticipating a 16%increase in most functional areas.This investment is in business continuity(asset protection),which has been driven by both political and technical concerns about the macro-societal environment,including post-COVID-19 health considerations,cyberattacks,natural disasters,and

111、wars in several regions.The public sector considers edge as more of a cost saving technology with 71%of public sector respondents expect a greater than 11%cost savings from edge.Value generation from edge is not seen as the primary driver for public sector as 62%expect less than 10%value from edge i

112、n terms of revenue.The number of edge locations in the public sector in the 50-99 range will increase 50%by 2026,with 70%of public sector customers expecting to invest less than$500K.In terms of maturity public sector show a cautious approach,but not one that has dismissed AI and edge.Only 4%of resp

113、ondents stated they have no plans,but only 6%have live AI deployments currently.However,47%of respondents are piloting AI use cases,which indicates that the public sector could have a majority of organizations with live AI uses cases within a couple of years.Public sector organizations report in Omd

114、ias Enterprise Insights survey 2023 that building a modern workplace was the top digital transformation initiative,with 13%stating this is well advanced,which is why public sector put improving employee productivity as its number two priority where AI and edge can deliver results.Figure 14:Public se

115、ctor use cases for edgeQ4:WHAT ARE YOUR ORGANIZATIONS GREATEST CHALLENGES/AREAS OF IMPROVEMENT(PUBLIC SECTOR)?SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)10%20%30%40%50%60%Product/service innovationFraud preventionMake complex data,intuitively accessible and actionableLarge scale data

116、 search,retrieval and analysisContent generation at the click of a buttonMarketing,sales,and customer engagementCyber security(e.g.,ransomware,malware)Fast process of customer dataPhysical security and asset protectionSustainability and environmental proceduresEmployee safety and trainingAutomate in

117、cident resolutionEmployee productivityRegulatory complianceSupply chain optimization0%60%50%45%44%41%36%36%32%30%25%22%20%20%17%11%?HOMEAN OMDIA EBOOK212024 STATE OF EDGE COMPUTINGTELECOMMUNICATIONS NEARLY 50%ARE FOCUSED ON CUSTOMER AND EMPLOYEE EXPERIENCEFigure 15 shows that for telecommunications

118、organizations Improving customer experience and cyber security are the top use cases for edge.This view was collaborated by the industry at its big global event Mobile World Congress(MWC)Barcelona 2024.The take-away from MWC was that the relentless focus on AI has reignited interest in the evolution

119、 of telco edge nodes and how these nodes can be monetized for third-party workloads.MWC was full of AI initiatives,most of which were still somewhat nascent and preparatory for the widespread usage of the technology in the future.However,what was clear is that AI is intended for telco networks as an

120、 enabler of better networks and third-party services.The evidence of the optimism in edge and AI from the telecommunications sector is evident as 25%expect to get more than 20%in terms of cost savings from edge.Meanwhile the uncertainty of the commercial model was also evident,60%expect less than 10

121、%in terms of new value revenue from edge.This uncertainty is not restraining telco investment,as 32%will spend more than$500K on edge by 2026 and across telco an expected 50%increase in the number of edge deployments in the hundreds is expected by 2026.Telcos reported that they are very active in th

122、e use of AI and edge,with 20%of respondents reporting at least one live AI deployment in one business function and 15%scaling AI across more than business function.However,the majority of Telco respondents stated they are investigating use cases,26%.The biggest restraining force for telcos is cybers

123、ecurity and risk,37%of respondents put that as the top barrier to faster AI and edge adoption.This agrees with Omdias Enterprise Insights 2023 survey where improving operational resiliency was the top business challenge in 2024.Figure 15:Telecommunications use cases for edgeQ4:WHAT ARE YOUR ORGANIZA

124、TIONS GREATEST CHALLENGES/AREAS OF IMPROVEMENT(TELECOMMUNICATIONS)?SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)10%20%30%40%50%Employee safety and trainingAutomate incident resolutionSupply chain optimizationLarge scale data search,retrievalFraud preventionPhysical security and asset p

125、rotectionCybersecurityMake data,accessible and actionableCustomer engagementImproved customer experiencesProduct/service innovationContent generationEmployee productivitySustainability and environmental Regulatory compliance0%48%47%47%46%40%36%35%30%30%29%29%19%19%13%6%?HOMEAN OMDIA EBOOK222024 STAT

126、E OF EDGE COMPUTINGUTILITIES NEARLY 50%ARE CONCERNED ABOUT CYBERSECURITYFigure 16 shows that cyber security is the top use case for edge in the utilities sector.Omdias ITEI 2024 survey found that work force mobility was the top use of AI in the sector.Which combined with its(utilities sector)extensi

127、ve monitoring and remote metering capabilities makes security an obvious concern and use case for its edge estate.However,this sector is not investing as much as other sectors in edge,with 33%expected to invest less than$100K by 2026,although the number of edge deployments in the hundreds will more

128、than double by 2026 to 24%of the estate.The utilities sector sees cost savings instead of new revue as the most likely source of financial benefit,with 69%expecting greater than 20%savings from edge.Meanwhile,73%expect less than 10%in terms of revenue value from edge.Utilities organizations are more

129、 cautious about the adoption of AI and edge,with only 11%of respondents stating they currently have a live deployment in at least one business function.The majority,43%,are piloting at least one-use case,with 29%still investigating possible use cases.On deeper analysis,utilities reported that risk a

130、nalysis and mitigation in real time is the most transformative capability to its business,62%ranked it most important.This agrees with Omdias Enterprise Insights 2023 survey where the top business challenge was increasing operational efficiency for utility companies.Utilities report that a lack of s

131、kills and security concerns are the two biggest barriers to faster adoption of AI and edge.These challenges demonstrate why utility companies report that more managed services options for AI and edge would be relevant to them.Managed services would help resolve the skills challenge and security issu

132、es by providing a trusted partner.Figure 16:Utilities use cases for edgeQ4:WHAT ARE YOUR ORGANIZATIONS GREATEST CHALLENGES/AREAS OF IMPROVEMENT(UTILITIES)?SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)10%20%30%40%50%Supply chain optimizationContent generationAutomate incident resolution

133、Employee productivityPhysical security and asset protectionFraud preventionImproved customer experiencesEmployee safety and trainingCybersecuritySustainability and environmental Product/service innovationCustomer engagementMake data,accessible and actionableLarge scale data search,retrievalRegulator

134、y compliance0%49%38%37%36%36%36%36%35%33%30%30%26%22%11%0%?HOMEConclusionAN OMDIA EBOOK2024 STATE OF EDGE COMPUTINGThe survey clearly shows that edge is a topic of interest and a topic that all organizations have plans for in the next 12 months,because 100%of respondents stated they have an edge pla

135、n for the next 12 months.Cybersecurity and lack of skills are the top two barriers to faster adoption of edge and enterprise customers want more MSP offerings for edge to make deployment easier(see Figure 17).All organizations agree that the ability to scale and manage standard software from cloud t

136、o edge,84%,is the top essential requirement from any edge solution provider.Overall use cases are very industry specific but are focused on process improvements or customer experience.This is why most organizations expect more in cost savings than increased revenue from edge.However,edge deployments

137、 in the hundreds are the fastest growing segment,which shows that edge is beginning to scale.This is reflected in the average investment in the edge,which is expected to be$740K by 2026,with enterprises spending nearly$1 million on average on edge.23Figure 17:What is needed to make edge easier to de

138、ploy?Q22:WHAT WOULD MAKE EDGE COMPUTING MORE RELEVANT TO YOUR ORGANIZATION?Enterprise Mid-market SOURCE:OMDIA(RESPONDENTS COULD SELECT MORE THAN ONE OPTION)Better cross-organization awareness of edge computingLower prices/more flexible contract termsBetter defined standards for edge computingMore ma

139、naged service optionsA larger marketplace of edge appsImproved management/orchestration toolsImproved code-app to develop edge apps easily0%20%30%40%10%50%60%70%40%35%42%49%45%36%62%53%55%54%57%58%36%36%?HOME24Appendix?HOME25AboutGoogle Cloud Google Cloud is the new way to the cloud,providing AI,inf

140、rastructure,developer,data,security,and collaboration tools built for today and tomorrow.Google Cloud offers a powerful,optimized AI stack with its own planet-scale infrastructure,custom-built chips,generative AI models and development platform,as well as AI-powered applications,to help organization

141、s transform.Customers in more than 200 countries and territories turn to Google Cloud as their trusted technology ?Google Cloud linkedin Google Cloud?Google Cloud?HOME26AboutOmdia Omdia is a global technology research powerhouse,established following the merger of the research division of Informa Te

142、ch(Ovum,Heavy Reading,and Tractica)and the acquired IHS Markit technology research portfolio*.We combine the expertise of more than 400 analysts across the entire technology spectrum,covering 150 markets.We publish over 3,000 research reports annually,reaching more than 14,000 subscribers,and cover

143、thousands of technology,media,and telecommunications companies.Our exhaustive intelligence and deep technology expertise enable us to uncover actionable insights that help our customers connect the dots in todays constantly evolving technology environment and empower them to improve their businesses

144、 today and tomorrow.*The majority of IHS Markit technology research products and solutions were acquired by Informa in August 2019 and are now part of Omdia.Author Roy IllsleyChief Analyst,Cloud and Data Center Research P?HOME27SurveyMethodology Omdia developed and conducted an independent research

145、survey of 640 organizations via a structure set of questions around the topic of edge computing and AI.The responses of these questions were analyzed by a team of analysts in the cloud and data center practice at Omdia,and cross checked with other Omdia data on the topic from both a horizontal as we

146、ll as an industry vertical perspective.The team of analysts generated a detailed presentation which was used to write this report.DemographicsThe 640 global organizations were 40%in North America and EMEA,and 20%from APAC.These organizations covered seven industry verticals:Banking and finance,Healt

147、hcare and life sciences,Manufacturing,Retail,Public sector,Telecommunications,and Utilities.57%of the respondents were classed as enterprise organizations with revenues greater than$1Bn.The survey was targeted at managers and directors in the technology division of these organizations,with 41%being

148、manager level,30%director level,26%VP or C-level,and only 3%individual team members.100%of respondents were familiar with edge computing and 86%were currently using public cloud computing.ReferencesEnterprise Insights 2023 n=4800(November 2023).?HOMEOmdiaE E W ?OmdiaHQ linkedin OmdiaCitation PolicyR

149、equest external citation and usage of Omdia research and data via The Omdia team of 400+analysts and consultants are located across the globeAmericasArgentina Brazil Canada United StatesAsia-PacificAustralia China India Japan Malaysia Singapore South Korea TaiwanEurope,Middle East,AfricaDenmark Fran

150、ce Germany Italy Kenya Netherlands South Africa Spain Sweden United Arab Emirates United KingdomCOPYRIGHT NOTICE AND DISCLAIMEROmdia is a registered trademark of Informa PLC and/or its affiliates.All other company and product names may be trademarks of their respective owners.Informa PLC registered

151、in England&Wales with number 8860726,registered office and head office 5 Howick Place,London,SW1P 1WG,UK.Copyright 2024 Omdia.All rights reserved.The Omdia research,data and information referenced herein(the“Omdia Materials”)are the copyrighted property of Informa Tech and its subsidiaries or affili

152、ates(together“Informa Tech”)and represent data,research,opinions or viewpoints published by Informa Tech,and are not representations of fact.The Omdia Materials reflect information and opinions from the original publication date and not from the date of this document.The information and opinions exp

153、ressed in the Omdia Materials are subject to change without notice and Informa Tech does not have any duty or responsibility to update the Omdia Materials or this publication as a result.Omdia Materials are delivered on an“as-is”and“as-available”basis.No representation or warranty,express or implied

154、,is made as to the fairness,accuracy,completeness or correctness of the information,opinions and conclusions contained in Omdia Materials.To the maximum extent permitted by law,Informa Tech and its affiliates,officers,directors,employees and agents,disclaim any liability(including,without limitation,any liability arising from fault or negligence)as to the accuracy or completeness or use of the Omdia Materials.Informa Tech will not,under any circumstance whatsoever,be liable for any trading,investment,commercial or other decisions based on or made in reliance of the Omdia Materials.?HOME