Defiance ETFshas launched a new exchange-traded fund (ETF) that combines two increasingly important layers of AI infrastructure—memory and optical connectivity—in a single U.S.-listed fund.
Defiance ETFs推出了一隻新的交易所交易基金（ETF），將人工智慧基礎設施中兩個日益重要的組成部分——記憶體器和光互連——整合到一隻在美國上市的基金中。
TheDefiance Memory & Photonics ETF (BATS:PRAM)tracks theSolactive Memory and Photonics Index, which selects the 20 largest eligible companies by free-float market capitalization and equally weights them. The index covers memory semiconductors, storage and networking chips, as well as optoelectronics.
Defiance 記憶體器與光子學ETF（BATS:PRAM）跟蹤Solactive記憶體器與光子學指數，該指數按自由流通市值選出20家符合條件的最大公司，並對其進行等權重配置。該指數涵蓋記憶體器半導體、記憶體與網絡晶片以及光電元件。
The fund's launch comes as AI infrastructure spending increasingly shifts beyond processors. AI models require enormous amounts of data to be stored, retrieved and moved rapidly between chips and servers.
該基金的推出正值人工智慧基礎設施支出日益從處理器擴展至其他領域之際。人工智慧模型需要海量數據被快速記憶體、檢索並在晶片與服務器之間高速傳輸。
Memory Becomes an AI Bottleneck
記憶體器成為人工智慧發展的瓶頸
The memory industry has emerged as a major beneficiary of the AI buildout, particularly through demand for high-bandwidth memory (HBM) used alongside advanced AI accelerators.
記憶體器行業已成為人工智慧建設的主要受益者之一，特別是對與先進人工智慧加速器配套使用的高帶寬記憶體器（HBM）的需求激增。
PRAM's portfolio includesSK Hynix Inc (NASDAQ:SKHY), Micron Technology, Inc (NASDAQ:MU), Nanya Technology (OTC:NNYAF), Winbond Electronics, SanDisk Corp (NASDAQ:SNDK) and Macronix International. SK Hynix has highlighted strong demand for HBM, AI server DRAM, and enterprise SSDs, while the broader industry is preparing for sustained AI-driven memory demand.
PRAM的投資組合包括SK Hynix Inc（納斯達克：SKHY）、美光科技（納斯達克：MU）、Nanya Technology（場外市場：NNYAF）、Winbond Electronics、SanDisk Corp（納斯達克：SNDK）和Macronix International。SK Hynix強調了市場對HBM、人工智慧服務器DRAM和企業級SSD的強勁需求，而整個行業也正在為持續的人工智慧驅動的記憶體器需求做準備。
That demand is also driving capacity expansion. SK Hynix has planned a $4 billion advanced packaging facility in Indiana, with HBM4E production targeted for 2029.
這一需求也推動了產能擴張。SK Hynix已計劃在印第安納州建設一座$40億的先進封裝工廠，並計劃於2029年投產HBM4E。
The opportunity extends beyond HBM. AI servers are increasing demand for conventional DRAM and high-capacity storage, creating a broader memory-cycle tailwind.
這一機遇並不僅限於HBM。人工智慧服務器正推動對傳統DRAM和大容量記憶體設備的需求，從而為整個記憶體器行業周期帶來更廣泛的順風效應。
Photonics — Another Side of the AI Infrastructure Equation: Moving Data
光子學——人工智慧基礎設施方程式的另一面：數據傳輸
LightCounting expects 800G optical-transceiver shipments to more than double in 2026, while 1.6T shipments are projected to reach tens of millions of ports from a small 2025 base.
LightCounting預計，800G光收發器的出貨量將在2026年增長一倍以上，而1.6T光收發器的出貨量預計將從2025年較小的基數增長至數千萬連接埠。
As AI clusters grow larger, electrical interconnects face limitations around bandwidth, distance and power consumption. Optical technologies are increasingly being deployed to connect GPUs, switches and data-center infrastructure.
隨著AI叢集規模不斷擴大，電氣互連在帶寬、距離和功耗方面面臨限制。光學技術正越來越多地被用於連接GPU、交換機和數據中心基礎設施。
PRAM's photonics exposure includesLumentum Holdings Inc (NASDAQ:LITE), while its broader semiconductor holdings includeMarvell Technology Inc (NASDAQ:MRVL), which plays a role in high-speed connectivity.
PRAM在光子學領域的持倉包括Lumentum Holdings Inc（納斯達克代碼：LITE），其更廣泛的半導體持倉則包括邁威爾科技（納斯達克代碼：MRVL），該公司在高速連接領域發揮著重要作用。
The combination creates a different AI infrastructure trade: memory stores and feeds the data; photonics moves it.
這種組合形成了一種不同的AI基礎設施交易邏輯：內存負責記憶體並提供數據；光子學負責傳輸數據。
A Global Supply Chain
全球供應鏈
PRAM also highlights how international the AI hardware supply chain remains. Eleven of its 20 holdings are listed outside the U.S., with significant exposure to Taiwan.
PRAM還強調了AI硬體供應鏈仍具有高度國際化特徵。其20隻持倉中有11隻在美國以外上市，對臺灣地區有顯著敞口。
Its top 10 constituents include Nanya Technology (5.79%), Marvell (5.68%), Sandisk (5.47%), Winbond (5.35%) Lumentum (5.35%), Macronix (5.25%), Jeju Semiconductor (5.22%), Micron (5.21%), Innodisk (5.21%) and SK Hynix (5.16%).
其前十大成分股包括Nanya Technology（5.79%）、邁威爾科技（5.68%）、Sandisk（5.47%）、Winbond（5.35%）、Lumentum（5.35%）、Macronix（5.25%）、Jeju Semiconductor（5.22%）、美光科技（5.21%）、Innodisk（5.21%）和SK Hynix（5.16%）。
With AI spending increasingly constrained by memory capacity, bandwidth and data movement, PRAM gives investors a targeted way to access two infrastructure bottlenecks that sit behind the headline GPU trade.
隨著AI支出日益受到內存容量、帶寬和數據傳輸的制約，PRAM為投資者提供了一種有針對性的方式，以佈局GPU熱門交易背後存在的兩大基礎設施瓶頸。 (Benzinga)