As　the　next-generation　internet　paradigm,　Web　3.0　aims　to　build　a　free,　equal　and　decentralized　Internet.　Different　from　the　data　monopoly　of　the　Internet　oligarchy　in　Web　2.0,　users　in　Web　3.0　are　fairly　endowed　with　the　right　to　store,　share,　access　and　manage　data.　However,　as　data　leakage,　tampering　and　loss　may　result　in　a　forfeiture　of　users’　data　control,　the　security　of　data　storage　has　become　a　key　and　essential　prerequisite　to　safeguard　users’　data　rights.　Blockchain　has　gained　widespread　acknowledgment　as　a　prospective　technology　for　solving　data　security.　Nonetheless,　its　primary　suitability　for　large-scale　datasets　is　limited,　owing　to　the　substantial　communication　and　storage　overheads.　Fortunately,　a　distributed　data　storage　network　(DDSN)　can　compensate　for　accommodating　large-scale　datasets　by　providing　sizeable　and　inexpensive　storage　space.　In　this　article,　we　first　summarize　the　data　storage　advantages　of　blockchain　and　DDSN,　and　analyze　their　security　challenges　in　terms　of　access　control,　authenticity　and　confidentiality.　To　address　these　challenges,　we　propose　a　general　and　versatile　data　security　storage　framework,　where　we　adopt　ciphertext-policy　attribute-based　encryption　(CP-ABE)　to　enable　user-centric　fine-grained　access　control　and　decentralized　authorization.　Furthermore,　to　reduce　storage　costs　while　protecting　data　authenticity,　we　propose　a　ciphertext　split　storage　model　for　CP-ABE.　In　this　model,　the　CP-ABE　ciphertext　is　divided　into　the　attribute　component　and　the　data　component　based　on　functional　structure　to　realize　on-chain/off-chain　split　storage.　In　addition,　by　embedding　the　ambiguity　factor　required　for　decryption　and　the　attributes　in　the　access　policy　into　the　index,　we　design　an　efficient　attribute-based　keyword　search　mechanism　to　resolve　the　conflict　between　confidentiality　and　availability　caused　by　encryption.　Finally,　we　demonstrate　the　effective　performance　of　the　proposed　framework　through　numerical　results.　IEEE